Comment author: ThomasSittler 22 November 2017 08:13:29PM 2 points [-]

Great work! This was enjoyable and useful.

Comment author: Kathy_Forth 11 November 2017 07:07:41AM *  5 points [-]

I suspect a lot of this is due to people trying to save time on reading. There are too many articles to keep up, so we (myself included) choose the articles that seem most likely to have the information I need most, and some of this priority order is author based. An additional method for people who are doing this for efficiency reasons:

We could do an experiment to find out what percentage of high status people's karma points are due to their status or getting a larger amount of attention overall than other posters. Then, us efficiency oriented people can mentally adjust the karma scores accordingly.

Comment author: ThomasSittler 12 November 2017 08:30:03PM 0 points [-]

You can also see the week's / month's most upvoted posts here. This gives you another way to filter for (hopefully) high-quality posts while losing only a small amount of blinding. (You know the ranking of posts that week, but you don't know their relative scores nor who wrote them).

Comment author: ThomasSittler 08 November 2017 09:02:27PM 3 points [-]

This paper contains some interesting results regarding how exactly one should aggregate credences.

Abstract:

How should a group with different opinions (but the same values) make decisions? In a Bayesian setting, the natural question is how to aggregate credences: how to use a single credence function to naturally represent a collection of different credence functions. An extension of the standard Dutch-book arguments that apply to individual decision-makers recommends that group credences should be updated by conditionalization. This imposes a natural constraint on what aggregation rules can be like. Taking conditionalization as a basic constraint, we gather lessons from the established work on credence aggregation, and extend this work with two new impossibility results. We then explore contrasting features of two kinds of rules that satisfy the constraints we articulate: one kind uses fixed prior credences, and the other uses geometric averaging, as opposed to arithmetic averaging. We also prove a new characterisation result for geometric averaging. Finally we consider applications to neighboring philosophical issues, including the epistemology of disagreement.

15

Modesty and diversity: a concrete suggestion

In online discussions, the number of upvotes or likes a contribution receives is often highly correlated with the social status of the author within that community. This makes the community less epistemically diverse, and can contribute to feelings of groupthink or hero worship . Yet both the author of a... Read More
Comment author: Michael_PJ 28 October 2017 01:36:51AM 22 points [-]

Easy money: https://userstyles.org/styles/150270/effective-altruism-form-anti-kibitzer

I'd tell you to keep it or donate it, but I want to encourage the norm that such offers represent a real cost, so I hereby commit to use this money entirely on hedonistic pleasures.

Comment author: ThomasSittler 28 October 2017 11:21:21PM 5 points [-]

Paid. :)

Comment author: ThomasSittler 27 October 2017 10:21:26AM 11 points [-]

As a concrete suggestion, could someone make a stylish (or other) extension that hides upvote counts and usernames on the EA forum? I would be delighted to use it, and would encourage others to do so, too.

Comment author: ThomasSittler 27 October 2017 05:53:28PM *  8 points [-]

I'll pay £35 (or equivalent) to the first person who makes such an extension (to my satisfaction) before the end of Sunday, 5 November.

Comment author: Kelly_Witwicki 27 October 2017 01:24:49PM 1 point [-]

If people have to opt into it, we can assume the people who currently misuse their votes won't.

Comment author: ThomasSittler 27 October 2017 05:49:32PM 2 points [-]

To clarify, I was suggesting this as a tool that anyone can use to be less biased. That seems orthogonal to the issue of people using downvotes to show disagreement (or whatever other misuse you had in mind).

Comment author: ThomasSittler 27 October 2017 10:21:26AM 11 points [-]

As a concrete suggestion, could someone make a stylish (or other) extension that hides upvote counts and usernames on the EA forum? I would be delighted to use it, and would encourage others to do so, too.

Comment author: Kelly_Witwicki 26 October 2017 08:49:13PM *  0 points [-]

I haven't thought about prioritization yet, and was hoping other people would discuss that here. Since a lot of these are actions individuals can take, it will vary a lot by what roles an individual plays and what they have the most room for improvement in.

That said... toning down jargon, I suspect you'd agree, is probably pretty cost-effective, as I would think is toning up the visibility of people from underrepresented groups. A Diversity & Inclusion Officer who could review and advise on social media communications, ads, community recruitment, website UX, conference speakers, talk content and descriptions, job postings and hiring processes, etc, and who could establish metrics and goals for and conduct annual reviews on inclusionary practices, sounds easily worth their salary, at the very least as an experiment for a year.

Comment author: ThomasSittler 27 October 2017 10:08:18AM *  4 points [-]

toning down jargon, I suspect you'd agree, is probably pretty cost-effective.

Could you link to some examples of EA content that you think are good at this, relative to the rest?

Comment author: ThomasSittler 26 October 2017 10:38:21PM *  3 points [-]

Thanks for the post. You propose a wide-ranging list of suggestions. Which ones do you think are most important, or would be most likely to lead to progress?

Depending on which causal mechanisms are leading to the level of diversity which we observe in EA, different solutions would be more or less effective. (Of course, culture can be a bit of a nebulous thing, so it's harder than usual to have a good causal model).

View more: Next