Beefy Boxes and Bandwidth Generously Provided by pair Networks
No such thing as a small change
 
PerlMonks  

(jcwren) Re: Quick and easy way to prevent multiple votes?

by jcwren (Prior)
on Jun 08, 2000 at 17:45 UTC ( [id://17077] : note . print w/replies, xml ) Need Help??


in reply to Quick and easy way to prevent multiple votes?

Heh. This about the time everyone starts thinking that maybe Intel didn't have such a bad idea with the processor serial numbers...

The problem with requiring an email address to vote may make it more awkward for internet cafe and handheld devices users.

No scheme that doesn't require a user to register is going to be full-proof. One of the determinations you have to make is how hard is someone going to try before it's not worth thier effort. As an example, the security on my system at work is tight, but only in the context that our users are naive. If someone fires up a packet sniffer, then my system can be compromised.

You also have to evaluate why someone would want to vote 300 times. Is there a cash reward at stake? Prestige? If they vote for Britney Spears 300 times, will her "music" get played that much more?

It can be hard to be objective about your site, because it's got your hard work in it. The site I manage at work they swore was going to need a T1 feedg, redundant servers, RAID, yada yada yada. The reality is if we get 100 hits a day, we'll be doing good. Now, it's targetted at an extremely specific market, so our information is important, but we're not running a site that's gonna get traffic like CNN or SlashDot.

If you can provide a simple method of preventing multiple votes, that may be all you need. Using the random number scheme mentioned above has some definite merit. Requiring accountability, however, is probably the best way. Just keep in mind who may be using your site, how often, and most importantly, WHY.

--Chris