I don’t think that can be done within the time limit we’re allowed. Each volunteer is given a stack of 50 petitions at a time and it take about 1 1/2 hours to go through the stack. Some have petitions have only one name, some have 10 or so. We do not have the time to check everything and compare them all.
It was done after the fact in the Kapanke (sp) recall and it turned out that there were not enough signatures to force a recall but too little, too late. He’s out.
Wisconsin needs to tighen the laws on recalls.
I have a very hard time believing this is a productive way ensure the integrity of the process.
For example, If I signed a petition every time someone asked me, my signature could be on many different sheets. It would only take 1000 frequent signers to add up to serious over-counts. How does your process identify those?
You also said addresses are not being verified. So If I signed, but each time I signed I used a different fictitious addresses you might only challenge it if it occurred on the same sheet?
What if I had my children sign it? any way to catch those?
In my opinion you guys need to hire a software engineering firm. I’m a software engineer and this is how I would tackle it:
1. Run each signature line through an OCR - Optical Character Recognition scanner.
2. Store all results in a database.
3. Flag anything that fails OCR as illegible.
4. Write a software “mashup” that does a look-up using Google or another mapping service to confirm each and every address. Flag all address fails.
5. Run a lookup against the State Department of Corrections list to flag all Felons. Most States have open web-services for this.
6. Run duplicate name look-ups, any Barry Buephane’s that signed 36+ times is a flag. Any John Smith’s that show up 2-3 times might have to be manually looked at.
7. Here is how you purge the list of Kids... You have to do a global “by address” count. Distinct addresses with more than 2 signatures should be looked at. The 3+ signature is likely to be a child. This is more effective with Democrats since they tend to not have large families.
What I am saying is the first pass through the data needs to be automated. You could have the software analyze the data multiple times using different rule-sets ranging from “Strict” to “Lenient”. If on the strictest settings the count comes out way about the threshold for validity then your wasting you time. However if on the most lenient rules the number of valid signatures falls short then it is time to dig in, call in the troops and start protesting signatures.
I’m honestly the RNC doesn’t already have a GO-Team up and running with a team of software engineers at the ready. RNC spends millions on their jet-setting chairman and leadership, certainly that would have anticipated how to deal with events like this.