Meta: notes on EA Forum moderation

The EA Forum moderators (currently Larissa Hesketh-Rowe, Denise Melchin, and Julia Wise) wanted to share some updates and clarifications on EA Forum moderation. Unfortunately the Forum FAQ isn't editable right now (looking forward to having a shiny new EA Forum, with editable sidebar, before too long!) so we’re just adding... Read More
Comment author: SiebeRozendal 02 January 2018 03:59:10PM *  1 point [-]

This is an interesting project! I am wondering how valuable you have found it, and whether there are any plans for further development. I can imagine that it would be valuable to

  • Increase complexity to increase robustness of the model, but then find some balance between robustness and user-friendliness, perhaps by allowing users to view the model on different 'levels' of complexity.
  • Use some form of crowd-sourcing to get much more reliable estimates, ideally weighted by expertise or forecasting ability.
  • Incorporate some insights from the moral uncertainty literature, so that low probability of something being very bad (e.g. wild animal suffering, or insect suffering) are given appropriate weight.

However, I have no idea how feasible this is, and imagine it would require many and valuable resources (lots of time, money, and capable researchers). Do you already have thoughts on this?

P.S. The link is missing for part IV

Comment author: Denise_Melchin 09 February 2018 12:55:44AM 0 points [-]

Thank you for your comment. I agree our model is only a very basic version and it would be interesting to see it developed further. (Though there are currently no further plans for development that I know of.)

This model was created in about 14 wks of FTEs. I expect a project like you're proposing to take much longer.


Causal Networks Model I: Introduction & User Guide

This is the user guide for the Causal Networks Model, created by CEA summer research fellows Alex Barry and Denise Melchin. Owen Cotton-Barratt provided the original idea, which was further developed by Max Dalton. Both, along with Stefan Schubert, provided comments and feedback throughout the process.  This is the beginning... Read More
Comment author: Kathy_Forth 11 November 2017 11:10:16PM 1 point [-]

"Multiple types of sex offenders exist. We may not have a complete list of different types yet."

This is a direct quote from the article, from a section covering a few different types of sex offenders. Section name: "Why we should not assume that effective altruism repels sex offenders"

I can't cover every single sub-topic in entirety in every single spot where a sub-topic is mentioned. The article would repeat itself a ridiculous amount.

I also cannot remove all mentions of all sub-topics that have not yet been fully covered. That would ruin all the natural connections inherent in the information. The article would seem to leave out a huge number of obviously important things.

This is why I support the implementation of a social norm where one doesn't argue with an author until after they've finished the article.

Comment author: Denise_Melchin 11 November 2017 11:21:00PM 3 points [-]

I agree that is a sensible norm. I'm sorry I implied you personally think that, I'll edit my comment accordingly.

However, since many people will stop reading before the article ends I think it's important to not get people get away with the impression this is what you think.

Comment author: Kathy_Forth 11 November 2017 11:02:41PM *  -1 points [-]

This post is long because:

There are a lot of myths and misconceptions about sexual violence. To have accurate ideas about the effective altruism potential of sexual violence reduction as a cause, one needs to be informed about a bunch of things at once. Given the complexity of the issue and the number of common misconceptions, a long length was the only way to do this topic justice.

This is a foundation article. Now that it exists, a series of short articles can be written based on the information and context contained in it to help raise awareness.

Comment author: Denise_Melchin 11 November 2017 11:17:41PM 4 points [-]

I don't disagree with the full content being laid out. I'm glad you wrote such an in depth article.

Although I think it would be better if you created an index of contents and having split it up into a few posts would help as well.

Comment author: Denise_Melchin 11 November 2017 10:47:51PM 4 points [-]

By the way, I think the below sentence is slightly wrong?

The ratio of male rapists to women outside the movement is around 1:8 (3:25), based on a 50/50 gender ratio.

Shouldn't that be 3:50 (in a group of 50 men and 50 women, you expect 6% of 50 = 3 male rapists)?

Comment author: Denise_Melchin 11 November 2017 10:43:23PM *  11 points [-]

Hi Kathy,

as I said before, thank you very much for your research into this! I agree with you that it is an important issue.

I'd be interested to hear about estimates of how much sexual violence lowers quality of life compared to other issues like poverty and depression. My hunch is that it causes similar amounts of suffering (whereby similar means 'within an order of magnitude') but I don't have any evidence for this.

Unfortunately your post is somewhat long which makes it a bit hard to read. More structure and maybe splitting it up into a few posts would help. People, even EA forum readers, tend to be lazy - and it'd be disappointing if thereby fewer people get informed on potential strategies to address sexual violence. I'm happy to work with you on this if you like.

I have to admit that I haven't read your post completely yet myself, so therefore I'm only commenting on one point related to the content for now.

I disagree with your characterisation of people who commit rape. [Edit: Kathy actually doesn't mischaracterise this in the article, but since it's a common misconception which is important to avoid I'll let the rest of the comment stand.]

I think the idea of the dichotomy of 'rapists' and 'non-rapists' is fundamentally misguided and a case of fundamental attribution error.

The type of rapist which you describe in your post who consciously makes the decision to ignore lack of consent isn't the only type of person who commits rape. Therefore the study you're citing only constitutes a lower bound of the number of rapists.

This is more speculative, but I think it's likely lots of people have done sexual activities with someone else without having sufficient evidence that their sex partner is consenting. It's only most of the time they get lucky and the other person wanted the sexual activities as much as them. If they're not so lucky, that makes them rapists.

That said, I do think there's a spectrum here - between people who cannot quite be bothered to properly check for consent to sex every time and people who will happily ignore lack of consent to sex in most situations.

Only addressing the latter of those can thereby only be a start to addressing the whole problem. This also leads to the important question of how the number of rapes committed is distributed. Are most acts of sexual violence committed by a select particularly egregious few or by the presumably more common 'casual rapist'? Answering this question is relevant for picking the strategies to focus on. This is because it seems plausible that different types of people who commit rape require different strategies to stop them.

Thank you for putting so much time and thought into your post.


Comment author: Denise_Melchin 25 July 2017 01:31:47PM *  0 points [-]

That's a great talk, thank you for it. This is why I've started to mind that people get encouraged to figure out what "their cause area" is.

Apart from the fact that they're likely to change their mind within a few years anyway, it's more valuable for the world for them to focus on what they're good at even if it's not in their preferred cause area. Cooperation between cause areas is important.

(Also, "figuring out what the best cause areas are" might be something that should also be done by people whose comparative advantage it is).

Comment author: Denise_Melchin 06 June 2017 03:34:29PM 3 points [-]

What is your model of why other people in the AI safety field disagree with you/don't consider this as important as you?

Comment author: MichaelPlant 07 December 2016 11:22:50AM 1 point [-]

in response to your first point, yes I did mix those up.

And for the 2nd, I'm thinking hedonically and am leaning on the literature on hedonic adaptation. I'm not sure how to think about re-doing the calculations if I was using preferences util. So I think it's consistent to say "I would give up much more than a year of life to keep my child alive" whilst recognising that few (any?) events have a long term impact on happiness, either positive or negative.

Comment author: Denise_Melchin 08 December 2016 06:42:41PM *  3 points [-]

I think the results on 'hedonic adaption' are much less straightforward than you think they are. In general I'd caution against making strong claims that completely go against common sense about people's preferences based on just reading a few studies.

View more: Next