Git history suggests chezmoi is still heavily concentrated around one contributor. Is that actually true?

[SushantVerma7969]

Hi,
I've been validating a Git-history analysis tool against real OSS projects and ran it on chezmoi.
One result surprised me.
Despite 5,700+ commits, 230+ contributors, and years of development, most major directories still appear highly concentrated in historical commit-touch data, with Tom Payne accounting for the vast majority of activity.
My question isn't whether the tool is right.
It's where this kind of analysis becomes misleading.
For maintainers and long-time contributors:
Does this roughly match the reality of knowledge distribution in chezmoi?
What important signals are missing from Git history alone?
If you were evaluating project resilience, what would you trust more than commit history?
I'm trying to understand where historical ownership metrics align with reality and where they break down.
Thanks.

[twpayne]

Yes, this is the reality for many open source projects.

Non-commercial open source projects typically have a primary maintainer, a handful of very-much-appreciated regular maintainer/contributors, and a very long tail of one-shot contributors.

I'm particularly appreciative of all the work that Austin, Braden, Felipe, Gal, Ruslan, and Zach put into this project.

The code contributions that you see in git commits alone are not a sufficient measure. There's also user support, code and design review, issue triage, evangelism, supporting software, and tens of other jobs that are a critical part of an open source's project success.

If you're trying to analyze a project's structure from its git history, then I strongly recommend reading The Mismeasure of Open Source by Andrew Nesbitt.

[Lockszmith-GH]

Thanks for the name-drop, really flattered.
There is something I was personally very curious about, GitHub introduced discussions circa ~2020, and this project started 2018.
Looks like first discussion thread is from 2021.

Before discussions did you have any other forum community discussion place? And is there a clear difference before discussions were used to after?

My personal feeling about projects with vs without a discussion forum (whether within GitHub or outside of it) is light-years apart.
The communities thrive even when development is kept at a minimum.
I was just curious if that is indeed what you saw, as the maintainer of this project.

[twpayne]

Before discussions did you have any other forum community discussion place? And is there a clear difference before discussions were used to after?

No, and this was deliberate as I don't want to be "on call" for chezmoi with a live chat. Github Discussions are a reasonable compromise between a live chat and opening an issue. They definitely help.

[SushantVerma7969]

Thanks, this is very helpful.

One thing I've been discovering while building the tool is that commit-history concentration and knowledge concentration are related, but not necessarily the same thing.

For example, a scope may look highly concentrated historically, while in reality multiple maintainers can review, modify, and support it confidently.

Your point about issue triage, design review, user support, documentation, and community work is especially interesting because none of those signals exist in Git history.

I'm now experimenting with treating commit-history metrics as indicators of historical concentration rather than ownership or maintainership, and trying to understand where those signals remain useful and where they become misleading.

I'll definitely read The Mismeasure of Open Source.

Thanks for taking the time to explain the distinction