Beefy Boxes and Bandwidth Generously Provided by pair Networks
Pathologically Eclectic Rubbish Lister
 
PerlMonks  

Re: Proposal: eliminate down-votes

by sundialsvc4 (Abbot)
on Jun 04, 2012 at 15:26 UTC ( #974317=note: print w/ replies, xml ) Need Help??


in reply to Proposal: eliminate down-votes

Let me put my money where my mouth is and direct you to e.g. the following web site: http://blog.stackoverflow.com/2009/03/the-value-of-downvoting-or-how-hacker-news-gets-it-wrong/

Now, let me be quick to point out that the case being made in this article is against my own view:   the author, Jeff Atwood, argues that downvoting should be allowed and that it is important.   But he goes on to say, about two-thirds of the way through the piece:

The problem isnít downvotes, per se, but encouraging responsible downvoting. Thatís why on Stack Overflow, we do it this way:
  • Upvotes add 10 reputation to the post author.
  • Downvotes remove 2 reputation from the post author, and 1 from your reputation.

And of course, the real information value begins where the article ends, in the 39-so-far follow-on responses.

For example, Steve Sheldon chimed in that, “Every community I have been involved in, the process of downvoting has caused such destruction, it is not clear to me how anyone would treat it so casually.   By destruction, I mean rating wars.”

CHMike later said: “The current model relies on the assumption that popularity ‘up votes’ is an indicator of pertinence and validity.   But this depends on the competence and quality of judgment of the voters.  Weighting the votes by the voter’s recognized ‘competence’ will yield a more valid estimation of the true value of the subject.   This is how it works in the scientific domain.   It is worth to take in consideration because it has been validated through a century old darwinian selection process.”

Max echoed my opinion when he said, “The problem with downvotes is it allows people who disagree to suppress the opinions of others.   The whole upvote/downvote isn’t a measure of quality, it’s a measure of differences between the post and the person reading it.”

Dan Paluska turns to the animal kingdom when he says, “How do you teach animals new tricks?   With positive rewards only.   You ignore bad behavior.   http://www.nytimes.com/2006/06/25/fashion/25love.html

I can freely accept that the only apparent response to my posting was 30 (so far...) downvotes.   It may well be that this section of the forum is only frequented by frequenters.   But, contrary to any opinion that I might be harboring puerile motives here ... I think that this is a very legitimate topic for discussion here at PerlMonks in the manner that I actually intended it (twice).   Right now, the system offers both up- and down-votes, of equal weight and no adverse cost, and only has a limited “dog votes” defense against bots and sock-puppets (presumably) which are used to blanket-downvote one’s chosen enemies.   There are hundreds of discussion articles similar to the one at StackOverflow.   But I don’t seem to be being too successful at starting one here...

I’m rather disappointed at that, frankly.   “Are we doing it right, or wrong?”   Given that we are looking closely at aesthetic improvements to the site, should we not also look at other aspects of how the software functions in service to the community?   Should there be a “poll” offered to the community-at-large?   (Has one ever been done?)   Why is the only response that has so-far been offered here, to this I-think important question, “thirty downvotes?”   Is this merely, as I frankly argue that it just might be, merely an effort to suppress a dissenting and hence unpopular view?

What arguments or discussions from elsewhere on the Internet might you bring to the discussion that I am (still...) trying to launch?   Or should I just saddle up my pony and vamoose?


Comment on Re: Proposal: eliminate down-votes
Re^2: Proposal: eliminate down-votes
by Corion (Pope) on Jun 04, 2012 at 15:51 UTC
    But I donít seem to be being too successful at starting one here...
    Why is the only response that has so-far been offered here, to this I-think important question, ďthirty downvotes?Ē

    I downvoted your node because of the blatant disregard to the 20+ replies you got to this root post and your other root post. If you're not going to look at and address the arguments these replies bring forth, but instead clamor for a poll, I can only assume that is because the replies are not in favour of your argument and you hope that a poll might be.

    What arguments or discussions from elsewhere on the Internet might you bring to the discussion that I am (still...) trying to launch?

    Why should arguments elsewhere have bearing on how things are here?

    If you're trying to launch a discussion, maybe that discussion would benefit from you not just arguing a side but addressing the arguments of the other side?

    I feel that my reply to your first such post was not addressed, and still stands. I won't repeat it as you seem to be unlikely to read it this time around.

    Update: Tyops

Re^2: Proposal: eliminate down-votes
by jdporter (Canon) on Jun 04, 2012 at 16:53 UTC
    Thatís why on Stack Overflow, we do it this way

    Our "dog votes" does essentially the same thing: excessive downvoting threatens an XP penalty.

    By destruction, I mean rating wars.

    I'm not sure what he means by "rating wars", but I haven't seen any phenomenon here at PerlMonks I'd describe as a "rating war".

    The problem with downvotes is it allows people who disagree to suppress the opinions of others.

    It does no such thing. No one is ever silenced (leaving aside the issue of reaping). And anyway, those who disagree are as entitled to express their opinion as those who agree.

    How do you teach animals new tricks? With positive rewards only. You ignore bad behavior.

    The problem with applying this in an on-line forum is that -- as I said earlier -- it completely eliminates the (very useful) ability to distinguish between a node which has been ignored or upon which people simply don't have much of an opinion, and a node which people actively regard as "bad" in some way.

    The whole upvote/downvote isnít a measure of quality, itís a measure of differences between the post and the person reading it.

    Actually, it's not a measure of anything (except of itself). The problems come when people start making assumptions about what the node rep system means, or should mean.

    I can freely accept that the only apparent response to my posting was 30 (so far...) downvotes.
    But I donít seem to be being too successful at starting one here...

    Uh... what about the 50+ nodes posted in this thread so far, including 27 direct replies to your post? Stop being so melodramatic. It doesn't help your argument.

    limited ďdog votesĒ defense against bots and sock-puppets (presumably) which are used to blanket-downvote oneís chosen enemies

    I don't think that's what dog votes are intended to do. In fact, they would be particularly ineffective at doing that. First of all, only a high-ranking monk would have enough votes at his disposal to do any kind of carpet bombing of a chosen enemy; and such high-ranking monks are buffered against the effects of losing XP by the wide XP increments needed to gain (or lose) a level. For example, a mid-level Canon (26,000 XP) gets 38 votes to expend in a day. If he uses them all for downvoting, he'll lose (on average) 13 XP per day. At that rate, it would take him well over 300 days to drop down to level 19. And even though that would be somewhat shameful, the only concrete penalty is that he'd be getting 36 votes per day instead of 38. Even at 36 or 38 votes per day, using them all for carpet bombing, one person can't really do much damage to another monk's XP, except for the lowest level monks. And what high-level monk is going to carpet-bomb a newbie, unless he really deserves it?

    Is this merely ... an effort to suppress a dissenting and hence unpopular view?

    Unless your node is in danger of getting reaped, no. A post with negative rep is just as visible as post with positive rep. This is especially true in sections other than SoPW.
    Update: And this is in distinction to some other sites such as Slashdot, where a visibility threshold is applied, so that people don't (normally) see nodes with a rep lower than zero (or something).

    I reckon we are the only monastery ever to have a dungeon stuffed with 16,000 zombies.
      Also of note is that a single user can vote on a given node only once. This further limits the damage a single user can do to a node, or another user by proxy.
Re^2: Proposal: eliminate down-votes
by jdporter (Canon) on Jun 04, 2012 at 17:52 UTC
    Should there be a ďpollĒ offered to the community-at-large?

    I find it interesting (which is to say, ironic) that you argue against the capability for each post to be a poll, and then say you think a poll on the topic might be useful. Consider: your root post is a poll. The voting has shown that your proposal is unpopular. Yay, the system works as it is.

Re^2: Proposal: eliminate down-votes
by aaron_baugher (Deacon) on Jun 04, 2012 at 18:42 UTC
    I think that this is a very legitimate topic for discussion here at PerlMonks [....] I donít seem to be being too successful at starting one here...

    Don't sell yourself short; you've been very successful at starting a discussion. This post has been active for almost a week and has prompted dozens of thoughtful replies. It currently dominates the Weekly Best nodelet. I'd call that a successful discussion starter by any measure. To say that the "only response" has been thirty downvotes is to insult the people who took the time to write thoughtful responses.

    That you (apparently) haven't been successful in converting anyone to your way of thinking is a separate issue altogether. You seem to be saying that, as long as people disagree with you, it means that: A) you haven't couched your argument in the right way, or B) you've been prevented from reaching the right people in the right manner. So you restart the discussion in a new thread, perhaps stating your case slightly differently; and you suggest a poll, as if dozens of votes and replies aren't a clear poll in themselves. You refuse to accept that you could simply be wrong (which in this case doesn't mean morally or factually wrong, but simply out of step with the preferences of the rest of the group), and that manifests as a social obnoxiousness that grates on people.

    Or should I just saddle up my pony and vamoose?

    Honestly, if you can't be content at a forum that allows down-voting, and it's clear that this one will not be changing any time soon, then that may be your best option. I'm not trying to get rid of you, and I don't get the impression any else here is either. But there are other places to discuss Perl that may be more to your liking; there's no need to work so hard to remake this one your own way. Life is too short to spend your time on Internet forums that make you unhappy. And threatening to take your ball and go home....well, that kind of thing makes it pretty tempting to call your bluff.

    Aaron B.
    Available for small or large Perl jobs; see my home node.

      Extremely well analyzed and concisely stated. Thank you.

      Or should I just saddle up my pony and vamoose?

      I'd go beyond what Aaron said and emphasize that sundialsvc4 has been an outstanding contributor to the site (IMHO). I think it would be a terrible shame if he left over an issue as trivial as whether we should allow downvoting or not.

      I reckon we are the only monastery ever to have a dungeon stuffed with 16,000 zombies.
Re^2: Proposal: eliminate down-votes (assume)
by tye (Cardinal) on Jun 04, 2012 at 20:22 UTC
    defense against bots and sock-puppets (presumably) which are used to blanket-downvote oneís chosen enemies

    As was pointed out, nobody is using sock puppets nor bots to blanket-downvote you. Having investigated suspected sock puppets, suspected bots, and suspected blanket-downvoting, I can tell you that most of those suspicions turn out to be incorrect. (Blanket-downvoting used to be relatively frequent but not as frequent as people suspected and is now quite rare.)

    You don't even appear to have a sworn enemy.

    Despite tons of replies, you don't seem to have taken much note of them. You don't appear to have reconsidered your proposal in light of the replies before proposing it again. Beyond that, it seems you should re-examine your founding assumptions that lead you to your conclusions.

    - tye        

Re^2: Proposal: eliminate down-votes
by BrowserUk (Pope) on Jun 04, 2012 at 22:39 UTC

    Upon reading this post, I thought of much retort, thankfully, aaron_baugher said most of it more clearly and succinctly than I might have. So I'll confine myself to addressing this sentence:

    I can freely accept that the only apparent response to my posting was 30 (so far...) downvotes.

    That pretty well sums up why this place (and other sites; but that is their problem) need downvotes. A few more statistics.

    Without downvotes, your three posts in this thread would show: +8 +5 +1

    Which could only be read as a small but consistently positive support for your proposal.

    But the reality stands as -38 .v. +8 -20 .v. +5 -9 .v. +1

    Hence, a notional, small positive support turns into an overwhelming 82.7% rejection of your proposal, on the basis of the response to your posts alone.

    On the basis of pro .v. anti posts in the thread, it is nearer 95% anti your proposal.

    And on the basis of all the votes cast in the thread for posts that vaguely support you; versus those very clearly against, it is closer to 99%.

    The disenfranchisement of the ability to record an active rejection -- in favor of passive acceptance -- would place the monastery into a reality distortion field, leading to the same outcome as "a few good men doing nothing".


    With the rise and rise of 'Social' network sites: 'Computers are making people easier to use everyday'
    Examine what is said, not who speaks -- Silence betokens consent -- Love the truth but pardon error.
    "Science is about questioning the status quo. Questioning authority".
    In the absence of evidence, opinion is indistinguishable from prejudice.

    The start of some sanity?

      ++ on this particular post, and ++ on -- staying.

      Update: thanks BrowserUK for calculating the numbers, effectively presenting the results of a 'poll'.

Re^2: Proposal: eliminate down-votes
by Jenda (Abbot) on Jun 06, 2012 at 17:21 UTC
    How do you teach animals new tricks? With positive rewards only. You ignore bad behavior.

    And you end up with a fat spoiled brat ... I mean dog.

    If you really take a look at the animal kingdom, you'll find out that the animals use both positive and negative feedback to teach their offspring "new tricks". There's not just "here, that's what you do", but rather also "no! This is not how you should behave." But the animals did not read didactic texts from acclaimed professors with no actual students, child-less well-meaning feminists or mothers neglecting their spoiled brats 'cause they have no time for the kids due to all the conferences and books.

    Jenda
    Enoch was right!
    Enjoy the last years of Rome.

Log In?
Username:
Password:

What's my password?
Create A New User
Node Status?
node history
Node Type: note [id://974317]
help
Chatterbox?
and the web crawler heard nothing...

How do I use this? | Other CB clients
Other Users?
Others contemplating the Monastery: (9)
As of 2014-12-27 20:27 GMT
Sections?
Information?
Find Nodes?
Leftovers?
    Voting Booth?

    Is guessing a good strategy for surviving in the IT business?





    Results (177 votes), past polls