Because, at least in the case of CoCs in open source, the sorts of people most vocally pushing to implement and enforce them are almost invariably of the sort most likely to use them as political weapons and enforce them lopsidedly.
"Boot this developer from your project because they made a remark on social media I was offended by."
The problem here is not the CoC, it's that the organisation doesn't enforce the CoC, violates it, and punishes people who didn't violate it. They might as well not have a CoC in that case.
It just gives power to the accuser to say "Please refer to article 3 section 2 for compliant behavior. BANNED". But unlike a proper ruleset, the defense cannot do the same, because there's no actual definition of rude specified that I could say "In fact, I did none of those things, your honor"... and the accuser must iteratively prove his case. Instead I have to say "Im not being rude", and then it becomes a he said, she said, but with the accuser weilding "formal rules".
These are rules in the form of "don't be someone I don't like", and at best follows the fashion "the accused is presumed guilty until shown otherwise"
Besides the reams of examples of CoC's being abused and arbitrarily enforced like the one in the attached article you mean? I've never even heard of a CoC being applied in any way except a petty, bureaucratic way.
You don't hear about the system working right because it's inherently non-controversial.
The main difference I see is that the CoC also spells out enforcement actions and processes, while the enforcement process for the guidelines is whatever dang feels like it is. In other words from my perspective the CoC is less arbitrary.
An CoC on the other hand is exclusively about removal of members to achieve additional goals set by the CoC. It identify who is an "other" and gives power to a small group to opaquely remove the individual without constraints or liability. It also usually supersede any existing rules, goals and processes already existing in the community.
The distinction is thus in purpose, methods and results. If we imagine that we replaced the articles CoC with the HN guideline, we can read this: "When disagreeing, please reply to the argument". The Code of Conduct Enforcement Team could have talked to both Jeremy Howard and Joel Grus and steered the discussing on the factually arguments in favor or against the use of a Jupyter Notebook. They could have reminded both speakers to assume good faith, respond to the strongest plausible interpretation, and avoid unrelated controversies and generic tangents. It could remind them to focus on the purpose of the community, such as having a curious conversations with thoughtful and substantive comments.
Instead they exercised the power granted by the CoC.
Source? It seems like your assigning motives to others rather than listening to what they have to say.
> Instead they exercised the power granted by the CoC. If you actually read the parent you'd know they explicitly did not follow the CoC. The CoC sets out a process very much like the one you described in which the enforcement team is actually supposed to have a discussion with everyone involved and get all accounts of what happened. The fact they ignored the CoC isn't a problem with the CoC, it's a problem with the human beings not following the process.
Dang could start banning people at random, that wouldn't mean the HN guidelines are bad, it would mean dang decided not to follow the rules and has the power to do so.
The goal of conflict resolution is not about determining who is guilty and measuring out punishment, but rather to deescalate situation and fixing problems before there is a need for someone to be punished. The best moderation are those that are not seen because intervention occurred early and without escalating something small into something big.
The primary evidence for this exist in the names and words used in the CoC. A Enforcement Team is distinctly different than a Conflict Resolution team. A group full of mediators is different from a group full of enforcers. Enforcement is a different word from moderation.
If a CoC were focused on conflict resolution and had teams full of mediators, and where enforcement only came afterward when the mediators has failed and there exist documented proof of failed deescalation, only then would we have a situation where a CoC and a guideline would be much more similar in purpose, method and result. In which case the CoC enforcement would likely be called under a different name.
Whether the guidelines are used to punish people who act a certain way or just as a way to document how to act so as to avoid punishment is a fairly fine distinction IMO and maybe a distinction without a difference.
Sounds like a classic case of selection bias.
Or just get rid of them
--
Governance is the evergreen crisis. Was part of an org that had a constitutional crisis over conduct not covered by existing bylaws. HUGE drama. HUGE effort to resolve. Basically a mock trial.
Other orgs need more accessible, more lightweight ways to resolve their own slap fights.
I think they're right on point with this. Law is incredibly complex and you're likely to end up with a very shitty sub-implementation of the law when building a code of conduct, be it in itself or the application & litigation procedure.
"Be nice to each other & follow the local official regulations" should probably be a valid c&c.
20 years professional work, first time Ive ever seen that.
Noped out of there as fast I could.
Now, As a white male, I’ve seen it going the other way much more lately. Walk into a place where it’s all about diversity and tolerance, but failing, because hiring for diversity means passing on experience. Or frankly interest in doing the job.
So spent months cleaning up messes. The entire time everyone is bashing white males, unless non-binary.
Good formalized set of rules and rule enforcement focus on conflict resolution and deescalation, assumes good faith and give benefit of the doubt. Most legal system does this, but also community rules like Wikipedia.
From the article I don't see any attempt of conflict resolution between Joel Grus and Jeremy Howard. I don't see any process by NumFOCUS towards deescalation and refocusing towards a common goal. The process seem to have been opaque, the accusation hidden, purpose unclear. The authorities (the Code of Conduct committee) not held accountable. As a result it seems that their explicit rules did not serve a purpose of creating a professional environment.
FTA: “I think Joel is great, and I know for a fact that he doesn’t mind being called “wrong” (since the call I checked with him).”
Because they will be used to lawyerize innocent people out of the organization by entryist grifters. That's what they're for, that's what they're used for.
In every arena I've operated this is always the case. Military, Big Company and on the streets. Not saying it's right, but it's extremely common (in my experience)
> Rules shouldn't be implicit in professional environments.
Totally agree. But again, everywhere I've been there are at least 2 sets of rules.
Maybe I am jaded by conforming to the 'It is what it is' mentality. In practice though, that has been my experience.
That's why it is more favorable in today's environment to have explicit rules like CoC. At least it brings people to the same understanding quicker.
It does carry risk though. When living in a world with more explicit rules, people tend to treat anything not in the rule to be OK, and ignore most implicit rules. That will rush more rules to be explicit, also, written poorly without broader debates and discussions during these rushes.
At least up to a point where some people clearly, but mostly not openly acted as if these unwritten rules don't apply to them. And these transgressions were not dealt with from management even when they became aware of it.
Was a shitty situation if one was trapped in such a bubble within the organisation.
Fast forward being acquired by a global corporation with an endless amount of rules and regulations. Nobody reads them. Everybody knows the important ones. Nonetheless nobody really cares in the global corp.
Exactly what I see happening in our part of this global org now starting. A sad situation.
If OP had just been informally ostracized for whatever they actually did that made them not liked, they likely wouldn't be writing this blog post. They might find a different community, they might learn the unwritten social rules, or they might never learn. But they wouldn't need to write a blog post attempting to exonerate themselves from the black mark of a "COC violation" and rationalizing how the system attacked them.
Because code of conducts makes it easier to harass people. If there are formal rules then all a harasser has to do is to find something out of context and then start harassing the person by starting a CoC breakage investigation. Doing stuff like that without a CoC would get you kicked out.
> Let the attorneys do their job
The foundation of labor law in this country was built thanks to out-of-touch culture “warriors” and their lawyers going back to the early 1900s.
Why NOT formalize rules to some extent? Having a hidden set of implicit mores is far worse. In fact, you can easily read this post as a pro-COC argument.
After all, the author is able to make a really strong case that he didn't really infringe on any part of the COC, and that the COC was poorly constructed to begin with. With no COC, he'd have to just be saying "I think Numfocus exercised bad judgement," which is a much harder argument to make since it's so subjective.
We're professionals. Rules shouldn't be implicit in professional environments. Explicit rules are easier to use, and they make it easier to hold the authorities accountable for misusing them.