Voting System

February 25, 2004 | View Comments (43) | Category: Version 2

Summary: How the first round of voting will take place for Version 2

For Version 2, the first round of voting is a community process. However, I am sure we can all see the faults that a community process would bring in regards to cheating. If there were no prizes to be won, then I would not place so much emphasis on the voting process, but there are some prizes to be won. So damn good prizes to be won. So in light of this I have to figure out a fair way for everyone to vote while making sure that some people do not vote multiple times. The only logical method that I could think of at the moment is that if you wish to vote you have to physically open up your email client and send me your vote. That way, only real emails can be used and only one vote per email. Now if you have more than one email, I can't stop you from making more than one vote. Does this sound reasonable? If not, any suggestions?

Trackback URL: http://9rules.com/cgi-bin/mt/mt-tb.cgi/159

Comments

Why don't YOU send the email out to your subscriber base?

Mark Fusco (http://www.lightpierce.com/ltshdw)

Hi scrivs, umm i dont know if youd want something like this, but its pretty simple and it blcoks double voting cos the script remembers the users IP.
Heres the link:

http://www.cgi-world.com/pollit.html

hope that helps.

-peace

Steven

I'm with Steven. If you have the time, there is a minimum of at least 16.31 gazillion polling scripts out there in every language. Write your own or download and install. The whole email thing sounds good from a start, but its a hassle for your users (doubt you would get half as many votes as if it were an online vote), then you have to manually do the tallying, etc.

Lance Leonard (http://www.solarfrog.com)

...blcoks double voting cos the script remembers the users IP...

Which works great if those who would double post send muliple email from the same computer.

This is really a serious issue - since you're giving out "damn good" prizes which hold monetary value. Best to be able to prove you kept possible cheaters at bay should the 2nd place finisher who thought they should've won holds you or your corporation liable.

BTW : Just out of curiousity, this contest is void where prohibited - right?

Mark Fusco (http://www.lightpierce.com/ltshdw)

[ot]
This sucks. I have a layout already made in photoshop, and I don't think I have the time to break it out into an actual page. Arghhh, hopefully I can, at least get something out there.
[/ot]

Bryan (http://www.gamecubecheats.info)

you could have them send in a urine sample and take a blood test to negate double posts :)

Bryan (http://www.gamecubecheats.info)

There's no way to completely insure that nobody can cheat. It simply can't be done. IPs cannot be considered unique people, as they change from time to time. For dial-up users, it's as easy as reconnecting to the ISP. Emails can't be considered unique people, because as we all know, anyone can register multiple emails.

Just use your best judgment, and know that someone who actually resorts to cheating simply can't be identified. Most Internet-based contests use a panel of judges, and this is why.

Chris Vincent (http://dris.dyndns.org:8080/)

I was working on a voting system on php a while back. The best solution I could think of is asking the voter to supply an email address, where a confirmation mail is sent. The vote and the email address is stored in a separate file until the voter follows the link in the confirmation mail, when the vote is validated and moved to another file. At the same time it saves the email address and the user ip and uses them to filter suspicious votes.

Unfortunately something else came up and I didn't have the time to finish it and tie all loose ends.

Pekka Heikkinen

Could do an open vote in forum and only count regular posters. :-)

Or require voters to do the vote on *their* websites... I don't think anyone will go to the trouble of designing a bunch of new websites just to stuff the ballot box.

The IP thing is useless. I can sign on from AOL and have a dozen different IP addresses each time I hit the site.

JC (http://thelionsweb.com/weblog)

#10

I might go the route of sending an email out to everyone who has written a comment here...If you use fake emails I will have a post later declaring if I will go this route so you can leave the real email when you post a comment.

Mark: The 2nd place winner doesn't have to worry about the voters, because the 2nd round includes the panel of judges and I am certain that small group can't cheat ;)

Scrivs (http://www.9rules.com/whitespace/)

#11

Hi. While it's cool to get everyone involved in the contest by letting them vote, I can see some potential pitfalls with the whole comunity voting thing.

When you get, say, 100 entries, the entries that get listed first in the "submission gallery" will have a distinct advantage. Let's face it, most people don't time to look through that many entries, and so will just vote for one of their friends' entries or for one of the first few they see.

As you mentioned, people can skew the voting by voting multiple times, or by having their friends vote for their entry. Errr… Oops, I just gave away my strategy… ;-)

Anyway, someone mentioned that the only method certain of impartiality is the panel of judges. But, if you're entering the contest, you have to realize that much of the value of the contest is drawing the creativity out of people to create a showcase for what can be done in CSS/graphic design/information design.

It's all fun and games, right? Right?

Travis Cripps (http://www.apparentmotion.com)

#12

I've never commented on this site before, but I'm an avid reader. I would contend that going the route of only emailing those that have commented on the site previously would be excluding a great many people who enjoy the content of the site and the ensuing discussion, but don't feel the need or desire to contribute.

Dan (http://www.fathomable.com/)

#13

Hey Dan, glad I could sucker you into posting a comment :) I do understand that this could leave out a great number of people, but you offer no alternatives. The only other feasible way that I see is not have the community vote at all and only use a panel of judges. I am trying to keep this open for everyone, but in the end I will have to make the decision and either get praised or (more likely) lambasted.

Scrivs (http://www.9rules.com/whitespace/)

#14

Perhaps we can keep the public vote and panel vote separate (i.e. two categories) and split the prizes as well. ^_^

Zelnox

#15

Zelnox brought up a great point. Why not have both public voting and judges?

You could apply a weighting algorithm to the voting, so that the public votes help determine the finalists, but not as much as the judges. Maybe a 60% judges, 40% public voting.

Or maybe reverse that. I'm not sure… Just a thought.

Travis Cripps (http://www.apparentmotion.com)

#16

Just require everyone to vote as many times as they like, that way only those who are really devoted will win.

In all seriousness, why not use a script that tracks by IP like someone suggested here? Sure people will spoof it, but that's another issue.

When it gets down to it, it's all a popularity contest when you get the general readership to vote... and by that I don't mean which design is better, I mean which designer has more friends willing to vote.

I mean really, I can't remember the last time I saw a web design / people's choice award go to someone who did really good work but wasn't a popular designer.

I donno maybe my work judging the SXSW award, the Webbies, the W3Remix and about 100 other web award shows has tainted me... I am ruthless with my scoring... I think I burned my "10" card some time ago.

Nick (http://www.digital-web.com)

#17

"Just require everyone to vote as many times as they like, that way only those who are really devoted will win."
-- Nick
The winner will be a very good programmer. Hehe.

Zelnox

#18

do I win? (not sure if that'd execute properly, I've never had to use fopen for a URL before....

JC (http://thelionsweb.com/weblog)

#19

)

not only do I forget to close the parenthesis on my comment, my php code gets removed. :-)

here it is again sans the gt and lt symbols

[? while (a=a) {
fopen("http://www.9rules.com/whitespace/voteiethingie.php?winner=me", "r");
}
?]

JC (http://thelionsweb.com/weblog)

#20

Alas, my fantastic streak of lurking had to come to an end, so I might as well start speaking up regularly.

Scrivs: Sorry I didn't offer any alternatives earlier, I was posting in the middle of a rather unimaginative lecture. After thinking it over, I'm inclined to suggest that filtering by IP address (and perhaps including an email confirmation) is the best idea.

Of course, I also generally tend to think people wouldn't want to cheat, so I may be way off.

I wouldn't be opposed to seeing two categories of judging, for public vote and panel vote, if it comes down to that. In any case, I'll be happy. I had a lot of fun building my entry, and learned a lot from the experience. So now I'm off to finalize it and its accompanying case study.

Dan (http://www.fathomable.com/)

#21

Dan -
IP address is useless. You're better off setting a cookie on the client computer and checking for that.

I run a forum site that has members all over the world. Prolly 90% US members. I can view the IPs of posters, and see who else has posted from that IP. Any given AOL member whose IP I click on will result in every other member who uses AOL. Ditto for a couple of other ones (verizon is the other main one IIRC, we have a lot of verizon members, more even than earthlink)

Even people on broadband don't keep the same IP address always... mine changes every couple of months. Not a big deal here, but if you're coding something more important than a simple voting thing for paul, don't even begin to consider relying on IP address as a method of identification.

JC (http://thelionsweb.com/weblog)

#22

Scrivs.. no matter what way the votes go, I had a lot of fun trying to build something, and I'd love to see how it's compares to the other submissions.

You could go all paranoid by doing the whole ip check, then let the users create a voting account where they have to login to vote, and top it off with an email confirmation...
And afterwards, make sure to make a nice script where you can filter out "illegal" votes that got through anyway ;)

Jadwigo

#23

Make people pay to vote. If people want to stuff the ballot box, they end up paying as much as they get back in prizes.

OK, more seriously, Zelnox's suggestion of two sets of votes is a good one. Make it like the Cannes Film Festival where there are the audience awards (open voting) and the jury awards (panel voting). This way, we also get the benefit of seeing if the "critics" agree with the "general public" on what is best.

wink (http://site-unseen.net/wink)

#24

There ya go! Reinstitute the poll tax!
This isn't a nation anyway... no one has the 'right' to vote.. Charge a buck a vote, and give the money to a randomly selected loser, not the winner (discourage anyone from stuffing the ballot box expecting to get their money back)

JC (http://thelionsweb.com/weblog)

#25

Hate to be the resident droopy sunflower here...

But, I have a couple of stupid questions

1. (randomly selected loser) - What if the ballot money surpasses the value of the winner's prize?

2. If you're having difficulty determining a voting method, do you go through the same thing in determining the method of a "random" draw?

3. If you are asking people to pay to be involved in this contest, how does that fall under the next phrase which usually follows "void where prohibited" - "no purchase required."

Not trying to rain on the contest, just have a knack for looking at things critically.

Mark Fusco (http://www.lightpierce.com/ltshdw)

#26

Hehe, it's okay Mark. Nobody will have to pay to play.

Scrivs (http://www.9rules.com/whitespace/)

#27

A little explanation, before I'm sure I'll get blasted for being paranoid -

This is not a simple, little happy contest that Paul is providing to the readership of Whitespace. Rather, this is (according to the footer below) a U.S.-based corporation designing a contest for a worldwide audience...

...and corporations are held accountable (well, there are a few exceptions, I guess) for everything they do, especially when it comes to receiving and spending money or items of value.

Just needs to be considered within that context.

I think Paul understands where I'm coming from.

Mark Fusco (http://www.lightpierce.com/ltshdw)

#28

So if it wasn't a corporation, but a single person doing the contest would that kill all the legal stuff?

Scrivs (http://www.9rules.com/whitespace/)

#29

1) Paul keeps the difference or distributes the entirety evenly to all participants
2)select participant as lucky_loser from entries where rand() limit 1 (try doing *that* in MS SQL. And if you can in a statement that short, please email me so I can rewrite a bunch of code at work)
3)Just because contests often have that does not mean contests must have that. Raffles, for instance. But there's a basic flaw in your argument here -- the people voting are not, generally speaking, the participants. You're paying a buck to have your vote count, just like some news programs have you call a 900 number that charges you a dollar to register your yes/no opinion on a topic.

I'm not seriously suggesting Paul do this... but it would certainly work.

JC (http://thelionsweb.com/weblog)

#30

Interesting point. Paul, are you actually a corporation or did you just use inc because it sounded better to you? When you were discussing on here it looked like the overwhelming response was that LLC was the way to go... but as an LLC you'd identify yourself as 9rules LLC, not inc

JC (http://thelionsweb.com/weblog)

#31

There, now everything is no longer 9rules and its under me. I don't have a company. I am a failure. My company went bankrupt and it is only me left. No LLC's or Inc's around here. No siree.

Scrivs (http://www.9rules.com/whitespace/)

#32

Probably how it should have been in the first place. But you should still link back to 9rules somewhere. maybe "Paul Scrivens (founder, 9rules LLC/inc)

JC (http://thelionsweb.com/weblog)

#33

Swift move there, buddy.

MArk Fusco (http://www.lightpierce.com/ltshdw)

#34

What do you mean, Mark? ::jedi hand thing:: it was always that way. This isn't the corporation you're looking for.

JC (http://thelionsweb.com/weblog)

#35

I was thinking Paul Scrivens, founder of the Web...oooops, Gore took that one.

Scrivs (http://www.9rules.com/whitespace/)

#36

Oh my goodness - all along, I thought this was Parallel Rules, Inc., easy mistake with all the discussion of next generation development and all.