3
\$\begingroup\$

As per the answer to this question, as linked by SSD in this answer, we know that the malus on downvoting was placed to avoid downvotes related to irrational dislike (of certain topics, answers, or people).

I have seen an answer to a meta question (can't find it) that the system needs to be robust enough to power through irrational votes, as there's no real way to avoid irrational votes without exercising control over voting.

Downvotes can 'grey out' an answer, and a quick rush of downvotes can result in an answer being deleted or not-read by most viewers, thus destroying a good answer. But a rush of upvotes can result in an answer being accepted, and other good answers not being read, to an almost similar extent

So, since both irrational upvoting and irrational downvoting can both harm the quality of answers;

a) why is the philosophy of relying on good votes to outweigh irrational votes not applied to downvoting

b) why is upvoting not limited in some fashion to discourage irrational upvoting in the same way downvoting is

c) Why is the emotional investment in good answers (enough to downvote the bad ones at a personal cost) assumed to be higher than the emotional investment in irritation, dislike, or other irrational viewpoints that result in irrational downvotes?

\$\endgroup\$
  • \$\begingroup\$ Related: blog.stackoverflow.com/2008/12/vote-fraud-and-you \$\endgroup\$ – Grubermensch Jun 19 '14 at 23:16
  • 8
    \$\begingroup\$ I'm curious why you feel this is more appropriate to ask on a specific site than on the generic Stack Exchange meta site. Seems like they'd give better answers, or at least no worse, and provide a wider context for a Stack-wide philosophy. \$\endgroup\$ – BESW Jun 19 '14 at 23:29
  • 2
    \$\begingroup\$ @BESW - I don't have personal experience of the upvote/downvote levels on other SE sites. So I thought i'd bring it here before I bring it there. \$\endgroup\$ – user2754 Jun 20 '14 at 1:53
  • 3
    \$\begingroup\$ I get that, but your questions are about the philosophy driving Exchange-wide mechanics which individual sites have no control over, more than about site-specific behaviours which arise from a particular community's interaction with those mechanics. Not saying it's wrong to ask this question here, just that I suspect you'll get much more complete answers with better context and history from meta.se. \$\endgroup\$ – BESW Jun 20 '14 at 2:07
  • \$\begingroup\$ @BESW - As long as the search function is exchange-specific, and i'm talking about a specific behaviour in rpg.se, this will still be useful to have here. I expect to eventually take this to se.com's meta as well. \$\endgroup\$ – user2754 Jun 20 '14 at 3:36
6
\$\begingroup\$

Downvotes can 'grey out' an answer, and a quick rush of downvotes can result in an answer being deleted or not-read by most viewers, thus destroying a good answer. But a rush of upvotes can result in an answer being accepted, and other good answers not being read, to an almost similar extent

There is a difference here, in that grey-outs (and obviously deletion) exerts a very strong negative pressure to read the answer. If an answer is improperly upvoted, it does push down the other answers in the question while they're sorted by votes, but they remain accessible and can also be sorted by age and activity.

As for increased pressure to accept a high-voted answer, I don't think this is as much of a problem as you make it out to be. If anything, I see more often that a very good answer isn't the accepted answer simply because it was posted after an inferior answer had been accepted by the asker.

a) why is the philosophy of relying on good votes to outweigh irrational votes not applied to downvoting

Because downvotes are poisonous, and upvotes are not.

b) why is upvoting not limited in some fashion to discourage irrational upvoting in the same way downvoting is

Voting, both up and down, is limited to 40/day and a little time-delay. Also, as described on the blog, Stack Exchange has some nifty algorithms behind-the-scenes that watch out for undesirable upvote patterns.

c) Why is the emotional investment in good answers (enough to downvote the bad ones at a personal cost) assumed to be higher than the emotional investment in irritation, dislike, or other irrational viewpoints that result in irrational downvotes?

I don't think this is the case. The penalties are reflective of the reality that downvotes are more powerful than upvotes.

\$\endgroup\$
  • 2
    \$\begingroup\$ I think your "Downvotes are poisonous, upvotes are not" line sums it up great. I'll try to remember that for future discussions on this topic. \$\endgroup\$ – lisardggY Jun 21 '14 at 5:15
4
\$\begingroup\$

Simply, that our upvotes are wanted very much, and there's no effective way, even in theory, to separate the rational and irrational components of our capacity to form judgements. Votes are the result of a large number of conscious and unconscious motivations, and the necessary role of almost all of those motivation in forming our capacity for judgement is recognised.

That's not to say that all reasons someone might upvote are wanted in the computational mix at the end of the day, but avoiding those motives is done via different mechanisms. In particular, questions likely to generate popularity contests are not accepted. It's not that such questions are wrong or bad or wouldn't generate a wealth of opinion that people might enjoy and profit from sifting through—it's that we are horrible, terrible voting citizens on those questions. Similarly, there is a limit on our votes-per-day. The warning that the limit even exists only kicks in if you start voting very prolifically, so only voters at risk of voting indiscriminately get gently reined in.

So upvoting is limited, but in a completely different way. Different mechanics for different dynamics and different system goals. Downvotes get rep maluses because it works, while what is presented to us for our voting pleasure is gated at an earlier stage in the process, because that works, and voting indiscriminately is very gently discouraged by capping votes-per-day (while also giving you a badge if you hit the limit, to keep it from causing overcorrection), because that also works.

It's worth remembering that much of the infrastructure of SE was developed incrementally, over time, based on a data-driven design philosophy. The odd corners of the system are there because a problem needed addressing and that was the effective solution.

\$\endgroup\$
4
\$\begingroup\$

In addition of the thins already saids, I want to point that if you irrationally upvote a bad answer, it gets more attention, so it will probably get downvoted if it deserves that. If you downvote a good answer it is less likely to be read, so its punctuation is less likely to be corrected.

\$\endgroup\$

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .