The Shadow of SourceCred

,
1

Request for Perspective

I’ve had some concerns about the technology we’re building over the past few days, and I wanted to share to get feedback on my assumptions and whatnot. My hope is that someone will come along and dissolve these concerns.

Is “fairness” just an epiphenomena?

Basically, I had this sense that, in general, “fairness” in cred scores is an epiphenomenon, and that what’s really happening to establish cred scores is the same old interplay of power dynamics, leverage, and politics.

The current discussion about adjusting cred weightings more fair for community cultivation work is another great example. What I see as occurring is a survival adaptation. We need the great work of our community cultivators, and this seems to be more of a means of appeasing that branch. Out of the 25 replies so far, not one attempted to even roughly quantify how much cultivation work value is missed by our instance. That would seem rather important if we were actually trying to establish a fair quantification of value for cultivation work, no? Note that the proposal is judged by the opinion of individual stakeholders and not by any quantitative benchmark indicating fairness or unfairness.

My point here is that cred scores are not optimizing towards fair quantification of value. If they are, it’s an epiphenomena. On a deeper level, cred scores seem to reflect what’s needed for the organization—the organism—to survive.

Cred weightings/allocations are useful or not, not fair or unfair.

What’s useful to organizational survival isn’t always humane…

I wonder if right now we’re in a honeymoon phase where due to our tight knit, insanely capital rich (thanks Protocol Labs) community, we have the luxury of resting in the belief that we’re building something that promotes fairness and truth around the value of contributions.

Under the hood, cred is really evolving game theoretic equilibrium of power and material constraints. We can think of “fairness” as social leverage, but it’s not the only lever in the system. And it’s not necessarily the most powerful. It’s possible we only have the luxury of being fair, virtuous humans because of our privilege in global civilization that allows us to be here in the first place, and to spend the time working on this. How well will this generalize to other organizations??

Is this just a momentary anesthetic for the woes of capitalist power dynamics?

The first thing this makes me think is that we’ve ended up in the same conundrum that we dislike about “capitalism”. If there’s a class of people in a community who’s work is taken for granted and does not have power over the politics of cred, they can (and likely will over time) be exploited, whether clearly so or subtly through a paradigm that becomes blind to the exploitation.

6 Likes
2

I think that these are interesting and useful questions. I’d love to riff on them with you on voice.

I am less concerned about these changes and would use less dire language perhaps when discussing them. I think that you’re probably right in saying this arbitrary cred re-weighting is not best long-term fix. Are these questions in scope for the creditor? I don’t know.

I see these changes as a short-term fix that also allows the community to assent together that we want to rebalance our Cred towards different types of contributions. There’s also a neatness to incentivizing developers by introducing them to the same frictions that other members might’ve had in having their contributions recognized. You might be right in seeing echoes of power dynamics, leverage, and politics, but I am heartened to see those levers pulled in favor of non-devs on a software project especially.

3

Hey Miguel. That would be great. I actually was just using the current cred reweighting as an example. I’m more so trying to point at a more general phenomena that could exist for instances in general. Apologies if it came off as an attack on the proposal.

1 Like
4

I’m not super involved in this project, but I saw this post in my summary email and just wanted to say that I think establishing a practice / culture of keeping an eye on shadow (something we hide, repress or deny) is extremely valuable.

My experience in neo-utopian communities over the past 6 years has taught me that attempts to create new systems of power and organization do not exist in a bubble, meaning we import all of our shit with us, including our wounds from existing systems of power. If individuals and the community as a whole are not doing some kind of shadow work - making the unconscious conscious - then we are more likely to blindly cause harm, and be unable to see & acknowledge it, because we think we’re “one of the good guys” and somehow infallible.

I genuinely don’t know how much of that is happening in SourceCred, I just appreciate that you’re bringing the question of shadow here @eeli :slight_smile:

5 Likes
5

It’s nice to have this topic. @yetifree makes a good call :pray:on diversity & the need for this signal.

I think you made a good point here :

And it really echoes the current discussion of the props channel in which we need to have more granularity. Every “prop message” should have a category emoji at the end to have a more accurate measure & discussion on how cred is flowing.

On the concern you raise, it’s important to keep an eye on this. It’s a good call on getting some metrics to assess the fairness when we make changes. But I would also “relax” :beach_umbrella: given the fact that :

  • it’s been an experimentation for only like 2/3 years…making things different on the long run requires time

  • we’re dogfooding so it’s key for us to experiment, to make bold changes and to experience failure in the process to be able to share this journey for other communities.

  • It’s still fuzzy in my mind but I think there are two separate levels when questioning fairness.There is the tool we’re building. There is also the use we have as a community. The tool must be robust to provide the fairness the community wants to have

  • the existence of this tool to automate a measure with a set of parameters that can be changed : is already an improvement in this “rewarding value” space. Let’s acknowledge and celebrate the small steps before going straight to the bigger steps :smiley:

2 Likes