Hello, fellow builders and open-sourcerers in the Open Social Web! I would like to chime in here in the Atmosphere Community forums (might cross-post to SocialHub in the Fediverse community) regarding what’s the current state of bus factors and people “getting hit by a bus”, especially in the gen AI era where (mostly unreviewed) AI-generated contributions and witch hunting from anti-AI purists (aka “pipebombing in the replies/DMs” in my own words) often cause developers to either slow down development or leave the community entirely. I want both honest and civil discussions of this sensitive topic without the parts that often burns everyone off the discourse.
How did we got here?
Let’s say it started with me asking Claude (link to original transcript in case anyone asking) months ago about if there’s pre-existing academic/scientific literature regarding the bus factor situation that esclated quickly into planning to contribute into the open-source sustainability discourse by surveying the current state of these within the aforementioned ecosystems. Another reason why I this exist is that there were history of in-fighting between protocol devs and even its users alongside mass exodus of devs and users for different reasons from moderation and community toxicity (see part one of Blacksky origin story series) up to the “paid audits a la Matrix”-related headaches (as Soatok mentioned in his blog).
I’ll be also upfront about why I’m the one asking this here: I have an very autistic special interest in community health and how trust structures fail, and my pattern recognition is struggling not catch them up across every domain it pops out. The bus-factor conversation in OSS sustainability circles is really a smaller version of a much older pattern throughout the history — small groups holding a lot of keys to the kingdom (it’s a CGP Grey reference if you insist, but I am mentioning The Dictator’s Handbook and selectorate theory here for futher readings since it explains the mechanism part and not because I think it just excuses anyone), and what happens when that concentration breaks down, whether through burnout, departure, or in-fighting. I’ve watched variations of this happen in places well outside tech, including in my country’s political history, and the mechanics rhyme more than people usually expect. So this thread isn’t really is the bus factor bad but closer to what does it look like when a young, still-forming protocol ecosystem hits this failure mode before it's built the immune system older communities eventually grow.
Agenda / Purpose
The agenda / purpose of this thread is to:
- discuss about the bus factor/burnout situation among builders and open-source devs of both ecosystems in the public and share field notes about it
- develop a (informal) community survery regarding this either in the form(s) of asynchronous survey forms, usage of
augurtoolkit from CHAOSS project + TruckFactor and anecdotal notes from fellow developers and publishing its results (this could be optional, but I do not intend to turn this research into a capstone project at later years in college for reasons) - Other ideas, anyone? Feel free to chime in as long as you follow the community guidelines here.
Although honestly, I will be simply documenting things at the moment (maybe over at my digital garden) based on my current capacity as I am focused on academics lately as a college student but I’ll be open for collaborative work if you do.