Communication, Collaboration, & Norms on Using Slack
Our lab has a very active slack! PennLINC’s slack channels are within the Psychosis and Neurodevelopment Section slack organization (”Penn Neuropsych” org). This organization includes hundreds of members and norms will vary across groups in this org. Here we outline several of the best practices and norms for slack use in PennLINC.
Why we use slack. Fun fact: slack is an acronym that stands for “Searchable Log of All Communication and Knowledge”. Nearly all projects in the lab involve teams of scientists; correspondence often contains incredibly valuable knowledge. Pairwise correspondence — including email and direct messages on slack — are not searchable and thus all info contained there is lost. In contrast, information on open channels is findable forever. Searching the lab slack is a major resource for lab members.
Why we should use open channels. Using open channels allows us to find decisions and conversations that may not seem important at the time, but often turn out to be. Furthermore, team members may be in a meeting, doing focused work, or on vacation; asking on an open channel allows a broader group of people to respond in a timely manner. As such, we try to keep all correspondence related to a project on a project channel. Use of direct messages in general defeats purpose of slack, and should be limited to personal/ non-science correspondence.
!Important!: No PHI. Protected health information (PHI; identifying information including participant names, addresses, etc) should only be transmitted using health-system approved means. PHI should never be placed on slack. In general, code is ok to post on slack but it is good to avoid posting data.
Norms on communication. Lab members keep different hours, especially with travel and remote work. Thus, we have decided as a group that it is ok — not worthy of an apology! — to send a slack message after normal business hours. However, to ensure that this is not intrusive, we recommend adjusting your slack settings to silence notifications after your typical working hours (“View preferences” > “Notifications” under your profile on slack). While the lab norm is to try to help each other and respond to questions as quickly as possible, the default assumption is that slack message is NOT time sensitive. All lab members are encouraged to block time on their schedule with slack silenced to allow for focused coding, analysis, and writing.
Types of channels. There are several types of channels on slack. The most common type is a “Project Channel” — see below. In addition to project channels, there are general channels for announcements, informatics, specific imaging modalities, and specific methods. When you join PennLINC, you will be automatically added to a set of ~20 of these general channels. The description is usually listed under “channel description”.
Project channels. Each new scientific project should have a dedicated project channel, with every member of the team included. Beyond the project lead and Ted, at a minimum the team usually also includes one of the senior scientists from the informatics team and also the second author who serves as the reproducabibuddy. Every project should also have an associated github repo on the PennLINC org, a notion notebook in the PennLINC org, and (usually) a project folder on CUBIC. All these elements — slack channel, github repo, notion notebook, cubic project — should have the same name. A common naming convention is LastName_ProjectName
.
Where to post questions. It can be confusing where to post questions on slack. In general, we recommend first posting to your project channel, as responses will be more easily found. However, sometimes expertise beyond the core project team will be needed. For technical questions beyond the project channel, #informatics is a good bet. However, note that there are several dedicated channels for specific common methods, including #gams and #spin_tests. For questions on CUBIC, there is a #cubic channel. For nitty-gritty questions related to a specific imaging modality, look for #pennlinc_{modality} (e.g., fMRI, dMRI, etc).
When to email Ted. Ted is a member >100 channels, so things can get lost. For specific tasks that take time & consideration – like manuscript review, grant feedback, career development quesetions, and letters of recommendation – Ted prefers email. It is ok to DM Ted for quick (“one-off”) queries, but for tasks that take time (manuscript review, letters, etc) please email him directly.
A note on privacy. Whether using slack or email, all electronic written correspondence has a poor history of privacy. If the content of your correspondence is not something you’d want a lot of people to see, please strongly consider a phone call rather than a written message.
A note on collaboration. Specific cases where you should talk to Ted ahead of time include:
1) Starting new collaborations outside the lab.
2) Asking a member of our lab to do something that is likely to be fairly time consuming.
3) Any effort that requires repeated rounds of input / questions to a member of another lab.
4) Meeting with other faculty members outside our immediate group/section. In general, Ted tends to prefer to be present for collab mtgs with other faculty. In general it is a also good idea to keep Ted (via cc) in the loop on correspondence with collaborating faculty outside our section.
Troubleshooting collaborations. In general, the most stressful situations in science often come from inter-personal conflict. Usually this can be resolved with direct and honest communication. In large collaborative efforts, some degree of mis-communication is inevitable – and does not reflect anyone “doing something wrong” – please don’t hesitate to reach out to discuss.