Thoughts on “Community”

I have been working in “community work” for over two decades, mostly as a volunteer, rarely as a professional. I have moderated bulletin boards, managed Open Source projects, am active in a sports foundation and was part in creating multiple events. Most of that has been in the context of Free and Open Source Software (FOSS), but I also managed music communities for a while. My current main project is the Rust project, so I use it as an example a lot. I would go as far as saying that following and looking at world-spanning networks is an interest of mine and that I have a huge privilege of being both in networks and positions where I can follow that interest. This means that I have formed some opinions which I frequently refer to, but would like to write down from time to time.

This page serves as a frequently expanded document of rough thoughts I have on social and community subjects. Each subsection stands for itself. I will add, remove, edit things constantly.

For what it’s worth: depending on your angle, I don’t expect any of this to be new or innovative. A lot of these things are researched. Many of them are also part of management classics.

It’s kind of a FGO: Frequently Given Opinions.

I don’t believe “the community” is a good model

There’s no “Rust Community”. There’s also no “Metal Community” or “Ruby Community”. You need to speak at least in plurals. “The Berlin Rust Community”? Maybe. The “European Rust Community”? Uff, tall order already. The model of a community as “one thing” leads to a number of common mistakes.

1. You may start believing that what you see is all there is. Seeing community as a whole block may inform your strategy badly, as some people you categorise in that community don’t feel addressed or don’t want to be considered part of it. While you could start modelling “subcommunities”, the model may become thin so quickly it cannot serve as a bracket. It also betrays a false hierarchy.
2. Every model has bias. A global “the community” model will easily bias you towards the locations you feel most comfortable in. Naming communities you interact with directly can serve as a conscious effort to debias.
   • This bias is often clearly visible if you know where to look: many industrial FOSS projects accidentally bias towards the US hard and even within it on Silicon Valley, favoring its timezones, styles of working and interaction styles present there.
3. Worst case, parts of the community may completely fall out of perspective. A not uncommon experience I had in the Ruby world was that the “Ruby community” was criticised for decisions/styles that made total sense from an asian/Japanese perspective. This leads to subtle problems that come dangerously close or cross into racism.
   • Another example here is how substantial groups are not perceived as existing, especially if they are non-English-speaking. There’s e.g. huge communities speaking Spanish, Portuguese, Chinese and Japanese in almost any programming language. Other communities may not speak English as first language, but their statements may be weighted by the standards of native speakers.
4. This makes a lot of well-intended projects fail: “the community” cannot be addressed - most problematically, you might address a problem a certain community does not have, currently doesn’t prioritise or does not perceive (yet). The most effective projects are local projects, with local members.
5. “Community norms” are always a blend. E.g. the Berlin Rust Community and the Cologne Rust Community are different, both great. Both may have connecting bits through their active members of the Rust project bringing in culture and both being in Germany. In certain areas, even multiple local meetups have different attitudes or might not even be strongly connected. Berlin organiser communities are well connected in ways you don’t perceive when not part of it. (hint: a ton of them used to be Rubyists)
   • Example: Seeing any individual or group as “Rustaceans” or “Rubyists” is almost always a mistake.
     • I’m playing with this a little: I give most of my Rust talks in a Ruby shirt. That isn’t random.
   • You don’t need to know all this. My conclusion out of this is that you need local people to do effective work.
   • For purely online groups, similar points apply.
6. But! Global community norms do exist and are useful, as a guideline for people seeking connections.

“The community” creates an image of uniformity that does not exist. This quickly leads to individuals not feeling represented or feeling ignored. It is miles away from any reality and therefore impractical. Effective interaction with community requires effort to understand them, everything else will lead to failures of communication. It is a question of a bare minimum of respect, especially when talking about communities you have no connection to.

Whenever you talk about communites, ask yourself “which one exactly?”. Otherwise, I prefer talking about “space” or “ecosystem”.

Community happens and “Community management” is a Red Herring at scale

Unless you have a very definable community (like a bulletin board) at hand, I believe that “Community management” is a problematic model. Especially in a professional context.

For two reasons:

1. A community manages itself. There will highly likely be someone in local group with more influence than yourself.
2. If you were to really manage a community, you would also be responsible for all the mistakes.

I believe this is the one thing that DevRel got right: you can manage relationships to communities. Which people you talk to. Which ones to boost. Which ones to maybe hire.

This isn’t a random observation: I’ve seen whole environments being killed by e.g. a vendor having strong rules about what “community” is allowed to happen and which not. Which Meetups were allowed to be presented on their blog and which ones not. Partnership contracts that gave you the right to run “community events”.

At the same time, people taking the stance that they manage a (global) community often end up at the receiving end of a firehose of budget and decisions requests. Unsurprisingly, because that’s what they communicated.

It is far more effective to trust and observe and to be flexible about assisting when asked for.

The final problem is that, in a business environment, you might end up reading a management book to improve your skills. Educate yourself in diplomacy and politics instead.

Companies have no “community”

Companies or other organisations talking consistently about “their community” is a flag. And highly likely a red one.

This means that they only see community by their definition. The organisation declares who’s in or out. They make themselves the only important party in this space.

Rough check: if you can replace “our community” by “our clients” or “our users” and no aspect is lost, you are probably seeing this. Especially in press releases.

Understanding this is crucial if you want to run a community-oriented company: you are a part in something, but peoples reasons to flock around you are highly individual. Speak to them more directly, this will help you and recognises them better.

Your blind spot is huge

From any individuals perspective, their view of any group of people is most likely “the network of contacts and backchannels”. This can already be seen in a circle of maybe 10 close friends: some are still closer and news travels faster.

Scaling this up to interacting with clusters of thousands of people, missing groups and drifts of 10 or 100 individuals is just normal.

That’s not a problem, but full overview is impossible.

Being aware of this help communicating: make yourself accessible over trying to pull in all the news. Make sure that you can remain reactive to things suddenly happening. I’m not only talking about negative things here. Not being able and ready to react on, support and boost positive things is also huge issue.

Especially in a growing space like the whole Rust environment, this means constant re-tuning and pruning of your information flow and approaches.

Finally, actively turning your eye towards the “outside” helps: “Who’s not here, and why?” is a question I invented to think about the program of my first conference, eurucamp. It challenges your status quo.

Abuse and harm exists and your strategy around it matters

This is a subject with a ton of depth, but I only want to give a rough glance. It’s also one where you should read a ton of material if you want to actively go into detecting and managing what is described below.

“There’s no abuse in the X community” is a dangerous statement. It exists, you haven’t been told yet. Even more so, harm is often inflicted due to people ending up in bad feedback loops, e.g. when working together in a team, but can’t find good methods of working together. Some of it is fixable, some not. Treating it as a fact of life that you can’t ignore (although we can work towards a better world!), allows you to prepare your reaction the moment you become aware: acknowledging that all of the above exists gets you past the stage of denial.

My experience of organisations and groups that have understood this is that they are rarely part of open scandals and take no enduring image damage: they often take a direct stance in naming issues they found and are an environment where consequences are taken to detect a repeat early. They usually own issues as happening in their space, responsibly. Safety is also a function of what happens when it isn’t there.

To take a more positive approach, I believe that especially our stance on harm needs some sublety. My experience of running abuse handling on over 100 meetups and 15 conferences is: if harm was unintentional and both sides ready for an apology, it is not uncommon that you become part of moderating the apology and the first step of healing. The experience to apologising to a (maybe unknown) person is also daunting to many, helping a (sincere!) apology to be expressed well is good for all sides. I think that’s a nice thing, come prepared.

Taking a positive stance on the work of abuse handling has helped me a lot.

top