Sunday, July 24, 2005

My Dream

I'm still brainstorming my post on Web2.0. I have a goal of recording everything in the world into a database. Everything from the weather to my sales and tips, right down to what people ordered, along with demographic information such as age and sex of the individuals who ordered it. I want to collect the data on how I did my hair that day, and how much water I drank. What did I eat and who did I talk to. There has to be a pattern and a correlation somewhere. When not a single person asks me for Heinz 57 sauce in the first 2 months, and then multiple people ask for it in the same day. When I sell a disproportionate number of coconut shrimp in one day, it must mean something. There must be something somewhere that brings this all together and looks at the patterns that emerge. I remember a story about WalMart's collected sales data, that lets them know that sales of strawberry pop-tarts increase seven fold when a hurricane is coming through. I want to be able to see that. Coco Sales soar on days between 75 and 87 degrees that follow a day of rain. Tips are higher on the second Tuesday of odd months. Weird shit like that cause I know it would happen. The semantic web brings this dream closer by labeling all the information for easy processing and cataloging. I've always thought that it is a waste for people to have to do work with data that is already in a computer. At work we print reports out of the Panasonic terminals and the credit card processing machines, then we enter that data right back into the computer in the back. It doesn't make any damn sense. A while back everyone touted the idea of the paper-less office, I think they finally shut up after they realized they did it all wrong and now they have more paper than ever trying to get the information out of one system and into another.
When the paper-less office was introduced it was all these companies saying ok put all of your data into my software. Sweet now all my sales data is in software from company X and all my customer information is in software from company Y. But wait what if I want to correlate my sales data with my customer data. Well that's easy print it all out, and get the interns to put it into a spread sheet. If company X and company Y had thought about it a little further they would see what is coming to light now. With all the new syndication formats, and ways of combining data (rss, atom, rdf, ajax, etc) we are starting to see companies realize that they can no longer lock your information into their product, cause you'll just go use someone else's.

No comments: