I've always thought it would be very cool to see how much freely available data on the internet could be mined by an automated, free-roving, intelligent agent. I've tried a few times to get started designing something like that, but I always try and design too large of a system the first time through. That process just causes me to get bored, and makes me want to move on to something else. I need to get into the habit of doing small, fast iterations. It's so hard for me to produce something that falls way short of my ultimate goal. I know I need to look at it like a step on the way to my grand idea, but I can't ever keep that in mind while I'm in the middle of something.
Posts
Showing posts from May, 2004
- Get link
- X
- Other Apps
In the search for improving data quality at our company, I found a government site that maintains a list of all the known populated areas in the US, and their proper names. The site is part of the USGS (US Geological Survey) agency, and it is called the Geographic Names Information System .
We can use this data to help us give the properly cased names for cities, and counties. We currently use a system by Vertex Inc. called Sales Tax for L Series, to get sales tax information for specific locations. A side benefit of this software is that we can do ZIP code to City/County/State mapping, and vice-versa. The only problem with this system is that all of the City/County/State names are in UPPER CASE. By joining that data to the GNIS data, we can properly case all of the names for free.