Relatedly, should we have a strong dispreference for upvoting (especially strong upvoting ) people who work in the same org as us, or whom we otherwise may have a nonacademic interest in promoting*? Deliberately soliciting upvotes on the Forum is clearly verbotten, yet in practice I know that I’m much more likely to read work by somebody else if I had a prior relationship with them**, and since I only upvote posts I’ve read, this means that I’m disproportionately likely to upvote posts by people who I work with, which seems bad.
On the flip side, I guess you can argue that any realistic pattern of non-random upvoting is a mild conflict of interest. For example, I’m more likely to read forecasting posts on the Forum, and I’m much more likely to upvote (and I rarely downvote) posts about forecasting. This in turn has a very small effect of raising awareness/attention/prestige of forecasting within EA, which has a very small but nonzero probability of having material consequences for me later.
So broadly, there are actions along the spectrum of “upvoting things you find interesting may lead to the movement being more interested in things you find interesting, which in turn may have a positive effect on your future material consequences” all the way up to “full astroturfing.”
A possible solution to this is for people to reflect on how they came across the article and chose to read it. If the honest answer is “I’m unlikely to have read this article if not for a prior connection with the author,” then opt against upvoting it***.
It’s also possible I’m overthinking this, and other people don’t think this is a problem in practice.
*(eg funders/fundees, mentors/mentees, members of the same cohort, current project partners, romantic relationships, etc)
**I haven’t surveyed others so I don’t know if this reading pattern is unusual. I will be slightly surprised if it is though.
***or flip a coin, biased towards your counterfactual probability of reading the article without that prior connection.
Relatedly, should we have a strong dispreference for upvoting (especially strong upvoting ) people who work in the same org as us, or whom we otherwise may have a nonacademic interest in promoting*? Deliberately soliciting upvotes on the Forum is clearly verbotten, yet in practice I know that I’m much more likely to read work by somebody else if I had a prior relationship with them**, and since I only upvote posts I’ve read, this means that I’m disproportionately likely to upvote posts by people who I work with, which seems bad.
On the flip side, I guess you can argue that any realistic pattern of non-random upvoting is a mild conflict of interest. For example, I’m more likely to read forecasting posts on the Forum, and I’m much more likely to upvote (and I rarely downvote) posts about forecasting. This in turn has a very small effect of raising awareness/attention/prestige of forecasting within EA, which has a very small but nonzero probability of having material consequences for me later.
So broadly, there are actions along the spectrum of “upvoting things you find interesting may lead to the movement being more interested in things you find interesting, which in turn may have a positive effect on your future material consequences” all the way up to “full astroturfing.”
A possible solution to this is for people to reflect on how they came across the article and chose to read it. If the honest answer is “I’m unlikely to have read this article if not for a prior connection with the author,” then opt against upvoting it***.
It’s also possible I’m overthinking this, and other people don’t think this is a problem in practice.
*(eg funders/fundees, mentors/mentees, members of the same cohort, current project partners, romantic relationships, etc)
**I haven’t surveyed others so I don’t know if this reading pattern is unusual. I will be slightly surprised if it is though.
***or flip a coin, biased towards your counterfactual probability of reading the article without that prior connection.