Deprecated: get_settings is deprecated since version 2.1.0! Use get_option() instead. in /home/vemw27zv035v/public_html/wp-includes/functions.php on line 5324
Institutionalized Entity Identification Problems : The Practitioner's Guide to Data Quality Improvement

Institutionalized Entity Identification Problems

December 13, 2011 by
Filed under: Data Governance, Data Quality 

It has been a while since I posted an entry – mostly signs of “busy-ness” in trying to wrap up projects before the end of the year. However, I did have an interesting experience recently with one of our customers, with whom we are working on developing a best practices guide for data governance.

For this customer, I was being provided with badge access to be able to get in and out of the building, and we had an appointment with security to have the badge created (as well as a bunch of other security-type things). As some of you might know, I never use my first name, but go by my middle name, David. However, since my driver’s license has my full name on it, I was told that my badge would have my real first name (“Howard”) on it, and if I needed to contact security for any reason, I would need to give them my real first name (which I *never* use).

Also, when they took my picture, they asked me to remove my glasses. I have worn glasses since third grade, and no one would recognize me without them. Yet here I am with my new badge, with a name I don’t use and an unrecognizable picture.

I noted that I never use that name, and the security person said that it made it easier for them to enforce the rule and make sure the names are identical to official identification (such as licenses). I said that it made it easier for him, but not for me, and he had an interesting response. He basically said that I only need to deal with a small amount of effort, but he would have to remember thousands of exceptions, so it is better for individuals to carry a small amount of pain than he having to carry a lot of it.

From a data quality perspective, I laughed at the fact that they have engineered an identity resolution problem into their application. But I also think there is a little bit of wisdom in his response – perhaps it is better to distribute the effort than concentrate it on a single point. Thoughts?


Comments are closed.