Did you read the actual description of the incident in question?

Sorry, was being somewhat sloppy - I meant to broadly wave at "I think sexual harassment is bad". More specifically could say:
- It's probably suboptimal ex post to make an unwanted romantic or sexual advance on someone, but it can definitely be reasonable ex ante 
- If it's predictably unwanted then it's probably bad ex ante
- if it's uncertain and the asker is in position of power / you are in professional setting / other person is esp likely to feel uncomfortable etc etc, then it's probably a bad idea ex ante
- If someone has a pattern of doing this and has received complaints about it then either they should update that they are bad at judging when this is ok and be really careful to steer pretty clear of this kind of thing
- The community should in general take repeated complaints of this kind of thing pretty seriously and not let it be brushed off with "but this person is good in X way", because some fraction of the time this is the tip of an iceberg of manipulative/abusive/harassing behavior, and is the best chance you'll get to catch it
- Even if you conclude it's not part of a pattern of nefarious behavior, it's still pretty costly to the community, and actions should be taken that provide high confidence it won't continue happening

Presumably there is some level of [mildness / accidentalness of misconduct ] and [strength of response] at which the correct community response is to say "that response seems a bit too strong for the level of misconduct" - do you disagree?  If that's the case I don't think it's that helpful to claim that it's generically bad to ever say that you think a response is too strong - it has to depend on the specifics of a case.

Edit to add: I edited my original comment to hopefully address these misunderstandings

Yep - indeed - I assumed it's obvious to everyone that it's a bad idea to make [things that are perceived as] unwanted romantic or sexual advances towards people, and that serious action should be taken if someone receives repeated complaints about that. 

The intentions of my comment were to give information that might be helpful + informative for people deciding how to best achieve a goal of something like "make the community safe and welcoming for people in general, and especially for underrepresented, vulnerable, or easy-to-make-feel-unwelcome groups". As a potential member of such a group, I was assuming my experiences are at least somewhat relevant.

Similar to comments by Emma and Jonas, I want to prevent people taking away what I would consider a misleading picture of Owen's behavior, because I think people having a misleading picture will make it harder to achieve the goal.

I think people might read from this post that it matches a pattern of predatory or nefarious behavior - which definitely does exist, and which I think should be handled very aggressively by removing people from the community, which I have advocated in other cases.  Whereas, I think the pattern in this case is social clumsiness, some misunderstandings, and some psychological hangups around scrupulosity and needing to confess attraction and receive moral validation. I think the latter pattern of behavior should be handled differently than the former, and I think it's very important information for people to know when deciding e.g. how likely Owen is to cause similar problems in the future, whether or in what way they personally want to interact with Owen, whether to ban him from particular spaces, what interventions could be undertaken to prevent problems like this from occurring, or what information they should take away from this situation as to what their response should be to someone who is displaying predatory behavior. 

 I continue to think that decisions whether to ban people from shared spaces should at least consider what their net effect was. I think that excluding Owen from the community in the past would have made it less welcoming for me and even specifically would have made me more vulnerable to harm from some situations that were more sinister. There was a particular episode (related to people who are no longer part of the community) where I was in a situation which was pressuring and manipulative and had weird power dynamics, and Owen's support and advice was helpful for me keeping my sanity and perspective and avoiding getting sucked in. 

Obviously my experiences don't erase the negative experiences other people had, and the algorithm of "ignore complaints of bad behavior if someone else says the person seems nice" is not a good one. But "get information from multiple sources to understand what overall pattern of behavior things were part of, and what the net effect on the community was" seems like a better algorithm than "consider only evidence about the worst things people did".

I would also add that as far as I can recall, everyone I've ever spoken with about Owen has had positive experiences interacting with him. This includes multiple other people in the same sort of demographic situation as me. This is obviously not conclusive evidence of anything, and there can be all sorts of selection effects, but is at least evidence that there are multiple other people with similar positive experiences to me.

[Liberally edited to clarify / address misunderstandings]

I assume it's obvious to everyone that it's a bad idea to make [things that are perceived as] unwanted romantic or sexual advances towards people, and that serious action should be taken if someone receives repeated complaints about that. I assume everyone agrees that "ignore complaints of harassment if a few people say they're pretty sure the perpetrator is a good person / they're a pillar of the community / their work is valuable / etc" is a bad policy. 

I assume everyone has a shared goal along the lines of "make the community safe and welcoming for people in general, and especially for underrepresented, vulnerable, or easy-to-make-feel-unwelcome groups".^[1]

As a potential member of such a group who has had significant interactions with Owen, I think I have information that might help people to pursue that goal more effectively. I assume one sensible way to make decisions that improve the welcomingness for particular groups is to ask representatives of that group whether a particular decision would make them feel more or less welcome. In the absence of general solicitation to that effect (at least with respect to this decision), I am giving my unsolicited opinion (which obviously introduces bias, but I'm thinking the value is mostly by pointing out that some type of opinion exists rather than thinking that anyone should make strong inferences about the distribution of opinions).

Here is the information I have that might be useful:

I have many interactions with Owen, since 2015 or so, originally as a much younger (late teens), new-to-EA, and female person, and in many situations where he had substantial power dynamics relative to me. I found him extremely respectful, kind, supportive, and empowering. He actively increased my intellectual and social confidence and made me feel safe, comfortable and welcome in the community. I think it would have been a big loss to me personally not to have interacted with Owen, and from my perspective he substantially improved the fabric of the community. For me, I expect that having Owen present in a community space would make it feel safer and more welcoming, even if I didn't know him already. 

More specifically with respect to safety and unhealthy power dynamics, there was a particular episode (related to people who are no longer part of the community) where I was in a situation which was pressuring and manipulative and had weird power dynamics. Owen's support and advice was helpful for me keeping my sanity and perspective and avoiding getting sucked in. This was downstream of particular personality traits and ways of interacting with people that would likely be helpful to other people he interacted with too, including: listening intently and taking people very seriously even when they're junior; presenting his opinions in a way that doesn't force them onto others; being generally calm, reasonable and gentle in manner.

Similar to comments by Emma and Jonas, I want to prevent people taking away what I would consider a misleading picture of Owen's behavior, because I think people having a misleading picture will make it harder to achieve the goal of making the community more welcoming.

I think people might read from this post that it matches a pattern of predatory or nefarious behavior - which definitely does exist, and which I think should be handled very aggressively by removing people from the community, which I have advocated in other cases.  Whereas, I think the pattern in this case is social clumsiness, some misunderstandings, and some psychological hangups around scrupulosity and needing to confess attraction and receive moral validation. 

I think the latter pattern of behavior should be handled differently than the former, and I think it's very important information for people to know when deciding e.g.:
-  how likely is Owen to cause similar problems in the future?
- what is the net effect of Owen's presence in a community space?
- in what way do they personally want to interact with Owen?
 - what interventions could be undertaken to prevent problems like this from occurring in the future?
- what warning signs indicate that someone might be at risk of causing similar problems?
- what should they take away from this situation as to what their response should be to someone who is displaying predatory behavior? (in particular, I think that if this was part of a pattern of sociopathic or predatory behavior, I think the response is significantly too mild, and I wouldn't want people to reproduce this response in that situation)

I am also somewhat worried that the accounts as stated might provide a slightly inaccurate picture of the events; I am particularly surprised about the claim that "Owen did not stop making repeated unwanted attempts at contact after being asked to do so". Based on my experiences with him, I personally would strongly predict that this was based on misunderstanding or confusion about what the other person wanted rather than Owen knowingly ignoring someone's requests. This is obviously not hard evidence, but if many people had similar reactions it might suggest there was either something wrong with the investigation, or with the framing of the conclusions, or with my and other people's ability to predict people's behavior!

Again, I obviously can't know how much harm Owen's actions caused other people. But if I was asked to give an overall judgement, in terms of:
- harm caused to people from interacting with Owen
- benefit to people from interacting with Owen
- how inviting vs unsafe community spaces feel for people, if they know that people are excluded on the basis of behavior like Owen's
then I would guess that it is substantially better to include Owen in community spaces.

More generally, I feel subjectively highly confident that Owen is broadly well described as:

• highly morally scrupulous, kind, and careful person
• would hate to cause someone to feel unsafe or uncomfortable
• thinks a lot about other people’s comfort and how to make them feel happy and safe,
• would only ever cause someone to feel uncomfortable by accident
• would make concerted + successful efforts to change if he was told something he was doing was causing problems for someone
• tries hard to view and interact with people how they would want to be viewed and interacted with
• acknowledges his limitations and takes careful steps to compensate for these 
• engages deeply and sensitively with different perspectives and understands that good intentions are not sufficient for avoiding causing other people harm

and is not well-described as:

• someone who is callous about other people’s needs and sense of safety
• puts his own needs ahead of theirs
• feels entitled to ask things of them
• isn’t interested in learning about how his actions affected people or changing his behavior
• views or treats people in a sexualized or objectified manner
• doesn’t acknowledge power dynamics, social and societal context, highly subjective nature of social interactions
• thinks that if they weren’t deliberately doing harm, and it seemed fine to them, then everything’s fine and anyone complaining of being harmed is just making a fuss

I am sharing this under a pseudonym for reasons like "not wanting this content to be something that comes up if someone searches my name in a professional context". I'm hoping I can confirm my identity to the EA forum team and they can attest that the factual claims about my demographics and interactions with Owen are correct.
 

1. ^{^}

   Possibly more controversially, and not pertinently to the rest of this comment, I personally think we should give non-zero weight to the impact on the wellbeing of a category of people something like:  "neurodivergent or otherwise socially unskilled people who have serious difficulties and/or mental health issues related to shame, social anxiety, and deprivation of social/romantic connection". My guess (which is anchored very strongly to my subjective experience of the intensity of suffering from these two kinds of harm and from talking to a small number of other people who have experienced both, and therefore could easily be changed by testimonials from other people who feel the intensity obviously goes the other way ) is that there is a significantly larger amount of suffering caused by social anxiety, shame and loneliness than there is by sexual harassment in the EA community. 
    
   Somewhat relatedly, I think:  there is a meaningful way in which the treatment of Owen here is plausibly unjust, namely: in terms of the amount of harm caused, and the level of vigilance that would be required to avoid that harm, I think Owen and people similar to him are being held to a much higher standard than we hold other people to with respect to different types of harms. That is to say, I think most people have done many things that are more blameworthy (in terms of the amount of harm caused, and the effort that would have been required on their part to avoid it) as the things listed here. I think if we banned everyone from the community who had caused this level of suffering with this level of avoidability, I'm not sure there would be anyone left. This kind of justice or proportionality is obviously not the only criteria on which decisions like this should be made, and there are good reasons to have strict policies around particular types of harm even if those policies sometimes lead to locally unjust decisions, but I think it bears acknowledging.