Jack and Pat
drawbridgej at sympatico.ca
Wed Oct 31 08:59:18 CDT 2012
John, I sent this late last night (via gmail reply button) and it got bounced for being too large. I'm sending it as a stand alone message in another email system. Hope it makes it to you. ******------------ Original response -------------------********** I agree it doesn't sound like the typical database. But as you said, there are a number of facts about people and their preferences, habits, toys, etc. With the volumes you have and the diversity, and the common key that can be used to relate these various facts, you can respond to questions, suggestions or the "who does X and Y and is between 20 and 35". Seems a bit of a dream to a direct marketer trying to sell anything from auto insurance to new homes or furniture or travel. Seems more cost effective than designing, running and using new survey results every time a new "who does X and W and ..." comes up. It sounds, and probably is, a bunch of surveys of people for various reasons. And someone has seen that the various reasons and the various surveys can be related to get a "pretty good approximation" of some new feature - not in the initial survey objective. Not sure how you describe it other than Subject, DateCreated, Total Records, Origin of the Data, Type of Questions/Responses (choose a value vs supply a value), then a list of fields/properties/attribute or whatever the "things" and some indication of the number of responses per field "thingy". Some detail about the individual fields and the response types would be very useful to anyone wanting to use the data/facts in different environments. Just some ramblings that I'm sure you've considered. Good luck.