9

Pablo_Stafforini comments on CEA is Fundraising! (Winter 2016) - Effective Altruism Forum

You are viewing a comment permalink. View the original post to see all comments and the full post content.

Comments (35)

You are viewing a single comment's thread. Show more comments above.

Comment author: Pablo_Stafforini 07 December 2016 01:12:07PM *  9 points [-]

While I disagree with Michael and don't think we should discourage EA orgs from posting fundraising documents,* I'm disappointed that his comment has so far received 100% downvotes. This seems to be part of a disturbing larger phenomenon whereby criticism of prominent EA orgs or people tends to attract significantly more downvotes that other posts or comments of comparable quality, especially posts or comments that praise such orgs or people.

__

(*) I work for CEA, so there's a potential conflict of interest that may bias my thinking about this issue.

Comment author: Ben_Todd 07 December 2016 08:42:34PM 7 points [-]

I downvoted because I found the tone negative and hyperbolic. It won't kill the forum. I think a good norm the community is to always steelman before criticising. This would make us more welcoming and constructive.

Comment author: Pablo_Stafforini 07 December 2016 10:08:39PM *  4 points [-]

I agree that, other things equal, we want to encourage critics to be constructive. All things considered, however, I'm not sure we should hold criticism to a higher standard, as we seem to be doing. This would result in higher quality criticism, but also in less total criticism.

In addition, the standard to which criticism is held is often influenced by irrelevant considerations, like the status of the person or organization being criticized. So in practice I would expect such a norm to stifle certain types of criticism more than others, over and above reducing criticism in general.

Comment author: Ben_Todd 08 December 2016 03:52:58AM 8 points [-]

I think we should hold criticism to a higher standard, because criticism has more costs. Negative things are much more memorable than positive things. People often remember criticism, perhaps just on a gut level, even if it's shown to be wrong later in the thread.

Comment author: Ben_Todd 07 December 2016 08:44:23PM 7 points [-]

Also there's now 14 comments debating this issue and 0 comments debating how much funding CEA should get, which is a vastly more important issue. So this comment derailed the thread.

Comment author: Evan_Gaensbauer 18 December 2016 11:06:27AM 1 point [-]

If the CEA is seriously seeking feedback from unaligned EA community members on how much funding the CEA should receive, I have all kinds of thoughts on this I'll write up as a top-level post when I get the chance.

Comment author: RyanCarey 07 December 2016 03:10:53PM 3 points [-]

Yep, it's helpful to emphasize that upvotes and downvotes should be allocated according to whether (as indicated when you hover over the button), you "found this useful", or "didn't find this useful", not based on agreement!

Comment author: Owen_Cotton-Barratt 07 December 2016 07:10:49PM *  2 points [-]

I know this is the stated meaning, and I usually think it's correct to act on. In some cases when usage deviates from this, though, I'm not actually sure that people are making a mistake.

I think that happens most often on short statements of opinion. In such cases, there's not much ambiguity about how useful the comment was (opinions are always somewhat useful but don't contain amazing new insights). It's more useful to get a cheap instant poll of how widespread that opinion is in the community.

Notes:

  • I'm not confident in this, but to the extent that it seems wrong it would be if we thought posting short opinions was generally unhelpful. (I'd find that claim more plausible of LW, but still dubious there.) Otherwise to convey the information about distribution of opinions lots of people need to post.
  • Separate buttons as Benito suggests below might well be preferable. In particular they'd avoid ambiguity of things like the case in hand, which will be mostly read as an expression of opinion but also gives some considerations for.
  • This "instant poll" effect is to my mind the strongest reason for having voting scores on posts be public anyway. Maybe if there were separate buttons only the "agree/disagree" one would get displayed, and the "useful/not-useful" would be used to determine display-order for posts.
  • I was going to down-vote Ryan's comment to express that I disagree ;) But then I noticed that it was unusually helpful that he'd raised the point explicitly as it made it easier to have this conversation, and didn't know what to do.
Comment author: RyanCarey 08 December 2016 01:18:17AM 4 points [-]

The primary role of the vote buttons is to create the incentive gradient that determines which comments (and commenters) we get to have. This is perhaps the most powerful tool we have for incentivising some types of commentary. So I think we should practically always vote according to what comments we want to exist. On the margin, I think everyone (including you, based on your last bullet!) should vote more on usefulness.

Comment author: Stefan_Schubert 07 December 2016 04:13:01PM 0 points [-]

Yes - perhaps the text (particularly of the downvote button) should be changed to something that clarifies that.

Comment author: Benito 07 December 2016 04:23:31PM 1 point [-]

I definitely use the down vote button a lot to expressive negative affect, which is strongly influenced by 'disagree'. Having separate buttons could be pretty awesome.

Comment author: RyanCarey 08 December 2016 01:10:40AM 2 points [-]

It would also be terrible UI to have two 'up' and two 'down' buttons with different meanings.

The nearest thing that would be feasible would be to have, as does LessWrong, a polling feature.

Comment author: Owen_Cotton-Barratt 08 December 2016 09:39:43AM -1 points [-]

You could give a little (3x3? 5x5?) voting grid: usefulness is one dimension, agreement is another. Users have the option of hiding one of the dimensions, and maybe this is the default.