Policy Review A proposal for changing the way votes are done

So we were talking about CAP in #Pokemon , and user gec brought up a pretty good idea that I think we should discuss!

While Voting in CAP is better than it has been in the past, there are still some problems with it. First of all, they create a span of 24 hours where there really isn't any interesting discussion on the forums. Secondly, there is still the potential for bandwagoning to play a major role in voting because everyone can still see each other's votes! The proposed solution is to use blind voting.

By blind voting, I mean what the old suspect test voting threads in Policy Review did. In these forums, when a user who isn't a mod / smod/admin, their post is invisible to all but mods/smods/admins. While this might unfortunately require a forum being made just for CAP voting, I think this has several huge benefits for us to consider. First of all, this mostly solves the bandwagoning issue (I'll hopefully assume that most mods are mature enough to make their own decisions). Secondly, this would let us keep discusison threads open for much longer. Instead of just closing discussion threads after 24 hours, we would keep them open while the polls are up, link to the polls in the OP + make a new post alerting people that the polls are open, and let people make arguments for what the best option on the slate is there!

EDIT: We could use redirects to make it so the polls still appear in the main forum btw!

I am entirely behind this idea, I feel it could solve bandwagoning not only at the beginning of polls, but especially towards the end when people start to think, "Well, this idea doesn't have a chance of winning so I'll vote for that other one." creating incredible bias towards one or two ideas.

I'd like this idea.

I've found myself in other polls just looking at what certain users voted - capefeather, Deck Knight and bugmaniacbob, and just voting similarly or exactly like them, without an inkling of what it entails, or even what the other choices on the slate were. I would not be surprised if others do the same.

Before any decision is made on this issue, I think it is very important that we determine exactly how much of a problem bandwagoning is. While this seems like a good idea in theory, I feel like it would be more confusing and difficult for people, and would serve to alienate them from the project. At least in my opinion, community participation is the most important thing to this project, and anything that would serve to threaten the amount of participation need to be looked at very seriously before we even consider implementing it. Now I can't really speak to how much of an effect bandwagoning in our current voting system has had, I am being completely honest when I think the end results of of recent projects have been great, and as such, I have a hard time believing that there is a big enough problem to warrant major changes. Maybe it really is, but I think determining that is by far the most important thing to consider, because if it ain't broke, we shouldn't be fixing it, especially when fixing it could have an important negative impact on that which is most important to the project.

That being said, if we determine it really is a major problem, AND we determine that we can do it without making CAP less friendly to the community, then I would put my support behind such change.

On a slightly different note, while eliminating bandwagoning would be great, I don't really see any benefit in letting discussion threads run on longer. The fact is, if the discussion is still going on, the thread really shouldn't have been closed yet anyways, and the poll should not yet be up. And even so, I feel that our decision on such an issue as this should be based on problems with voting alone, and not any potential minor side effects such as this.

The idea is optimistic and very cool, but the problem lies in the implementation.

No matter what, so far as I am currently aware, implementing blind voting requires that we abstract the voting process away from the main CAP forum in some manner. This means, most importantly, that CAP voting threads get reduced visibility and thus fewer votes. As a community project, I personally value the quantity of votes much more than the benefits that come with blind voting, such as removing voter bias and much of the chance for bandwagonning. As a matter of fact, the reasoning I present here is why we don't currently have blind voting implemented for CAP. Unlike the once-Smogon suspect testings, CAP polls during a CAP happen a lot more frequently and don't have fixed number of participants. This changes a lot, and is the root of the problem here. If a proper implementation can be found that allows us to achieve blind voting in the main forum, then this should be implemented without hesitation. Until then, I am content with what we have despite the overt problems therein.

Furthermore and quite succinctly, I tend to agree with jas that letting the discussion threads linger well after they've devolved into a convoluted mess of ranting, raving, and repetition is a bad idea. Those things really should only stay open as long as is necessary; having them open longer isn't valuable, it's almost always the contrary: problematic.

I'll also point out that blind voting would eliminate the ability to edit your vote, which nullifies the only advantage I can see with continuing discussion threads during polls.

I would recommend for the edification of everyone here to review the last Policy Review thread we had on this matter.

True, this was back in 2009 (I'm surprised it was that long ago) but the process still would have the same weaknesses.

I would have to be against blind voting, mostly for the reasons Rising_Dusk posted. In the end, there are advantages and drawbacks to making votes blind versus making votes visible.

The advantages of blind voting is that the polls remove bandwagoning to a large extent and would make the poll results more reflective of how the community really feels. The disadvantages of blind voting is that it makes the poll threads less "visible", as it's just one moderator posting a thread and then the community members can't even see their vote again.

Though we get better results, we also will probably get fewer votes in a blind polling system than we do in an open polling system. In a project where community involvement is arguably the most important aspect, we should not be doing anything to jeopardize community involvement. Sure, there might be a little bandwagoning, but it's not worth alienating prospective members of the CAP community. The person who's making his first postin a CAP 3 poll could be the TL of CAP 11 or something like that.

I'm not also convinced that bandwagoning is as prevalent as people think it is. I mean, there'll always be a few people who vote just because Deck Knight voted that way, but how many people vote only because someone else votes that way? Not only that, but people are allowed to post reasoning as to why they voted that way in the vote post, so if we make voting blind then people will not be able to see some of the reasoning as to why each selection is a good choice.

The worst problem is that blind voting disallows people from editing their votes, so if a person posts a compelling argument after the thread goes up or someone makes a great argument on IRC, then it won't affect the vote as much. A blind vote will also most likely introduce a second policy review thread that closes discussion threads as soon as the poll goes up. That could be disastrous to activity because it pretty much silences CAP discussion for 48 or even 72 hours at a time multiple times throughout a CAP.

In short, blind voting looks nice, but the drawbacks greatly outweigh the risks. Sorry if this post is incoherent it's kind of late :P

I might as well post my thoughts here.

I've felt for a long time that voter bias and other such things were one of CAP's biggest problems with regard to polls - we seem to have managed to deal with (or brushed under the carpet, can't remember which) the problems of alternate accounts and tactical voting, but the problem of picking and choosing depending on names rather than merit has always existed, particularly with regard to polls like stats and movepools where it is difficult for the hurried voter to compare between options.

I don't find the argument that it disallows editing of votes particularly convincing, as as far as I am concerned any inability to edit of votes is a plus - it makes it clear that you can't flip-flop between options or depending on your mood. You vote and you stick with it. I won't conceal that I am sick to death of floating voters. You may well say "oh but what if you change your mind or accidentally press the submit reply button" - well then, by all means feel free to PM a mod saying "can you change x to y" and all will be peachy. In fact, feel free to put that in the OP, if it makes you happy.

This is a great idea. But as RD said, it struggles with implementation. Personally I see reducing the number of casuals who come in as a good thing, but this is outweighed perhaps by the number of future contributors and possible exposure that we could lose. If we can implement a way of doing this that keeps the voting thread visible on the front page, then I'm all for it. Perhaps we could lobby to make the subtitles able to be edited by CAP mods, to advertise the current poll, or else make it so that once the voting thread is posted, no more posts can be made for the duration of that time (obviously this is an extreme example).

Saying that blind voting is a negative thing in and of itself is to ignore the very real problems that it solves, in my opinion. The only issue it has is with implementation, and judging by the other opinions in this thread, this may well be enough to keep it from seeing daylight.

Plus it's a bit of a faff.

Rising_Dusk said:

The idea is optimistic and very cool, but the problem lies in the implementation.

Click to expand...

This essentially sums up the issue. I've chatted about this with some of the other CAP mods (capefeather, Deck Knight, and Elevator Music, for reference). While the idea is good in theory, it would be difficult to do in practice. Read the posts presented in this topic and the previous PRC thread on why this isn't currently possible. Perhaps if forums in the future could facilitate such an idea, we could then revisit it.

In short: we don't have the technology. If you wish to add some other thoughts to this thread, feel free to PM a CAP moderator and we can see about re-opening this issue and unlocking this thread. However, I think we are pretty set on the decision that this isn't going to happen anytime soon.