Real World Data

My childhood was very vanilla, to put it nicely. Everyone I knew was either a Miss, Mrs. or Mr. and had pronounceable names like Shawn Johnson or Grace Karber. Imagine my surprise when I met a person whose name didn’t fit into the cookie cutter mold. And then another. Soon my narrow view of the way names worked was being changed. This was all fine for a human to comprehend, but as I’ve started diving deep into databases and how information is processed and used, it’s easy to see how a computer could get easily confused.

You can put us in a literal box, but we refuse to be put in a figurative box!!

At its most simple, my understanding of a name was “First_name Last_name,” two separate words, both capitalized. Then you can add Mr., Mrs., and Miss as appropriate. Then I learned about Ms., and there’s also Dr. but some people like to spell it out like Doctor, and what about if a person is a Mr. and a Dr.? Would their name get combined into Mr. Dr., or Dr. Mr.? Or would they have to choose? Out of the vast possibilities of prefixes to names, I’ve only listed five and already have run into a conflict.

Developers have a habit of trying to fit people into their data, when they should be trying to fit the data to the people. This is how you build the user experience, and the trust that comes along with it. When you force someone to choose between Male and Female to sign up for your website, there’s a good chance you’ve already offended and lost a portion of your user base. Facebook got around this hurdle by expanding their options to include “Other,” and then allowing for multiple options to be selected. Not only does this give Facebook more robust data, it lets their users build a relationship with the site. By letting users fill in the blank, so to speak, they are allowed to express themselves on a deeper and more personal level than a radio button allows.

One way to marry mental schema and database schema is to allow for guided selection — to have a huge list of options that narrows the more input it’s been given. When a user looks for the high school they graduated from to add to their Facebook profile, they start typing into the selection box and are given a list of schools based on their input. It’s like autocomplete for databases, and it helps with data in the long run. Facebook keeps their options flexible, and if they need to, they can lock their data down later. It’s hard to do the reverse, after that exclusive reputation’s been built and no one wants to use the app anymore.

While the thought of allowing users to input their own “dirty data” probably makes some developers cringe, it’s important to remember that people aren’t edge cases — and that they push back on apps that treat them that way. While it may look great in the data to limit gender to a binary, male or female, that app probably won’t survive compared to an app that is less limiting. Data doesn’t have to be for analysis. Sometimes, it can be for sheer expressiveness. If that doesn’t bridge the gap between computers and people, I don’t know what does.

“I think I only truly understand how complex the real world is when I try to model it in a database.” -@pamelafox

One clap, two clap, three clap, forty?

By clapping more or less, you can signal to us which stories really stand out.