Thursday, January 23, 2014

Database?


     What in the world is a database?  I will use this blog entry to attempt to answer this question, mostly for myself.  The usually fairly wise Wikipedia defines a database as "an organized collection of data." Well, that much was pretty self explanatory and doesn't help my problem much.  It goes on to tell me that "the data are typically organized to model relevant aspects of reality in a way that supports processes requiring this information."  What? After a little more reading, I just decided to understand it as just a collection of information. What particular information depends on who is doing the collecting and what they record and save.  What comes to mind for me is a collection of demographic information of customers.  The names, addresses, phone numbers, whatever else comes to mind for all the past customers of Floyd's Auto Service for example.  He could use that information when it comes time to send out advertisements and such.  That is just a very simple example though, how about Amazon's customer database.  That information will need some way to sort it to be useful.
   Cue the database management system.  As I understand it, the DBMS is computer software designed to collect data, analyze it, and allow human users to interact with it. I find lots of lists of supposedly well known versions of these, so there are many options likely depending on the type of data and the methods of retrieval required by the individual situation. Databases even seem to need to access each other on occasion.  From what I understand, something called an open database connectivity driver allows all these different DBMS systems to converse.
     It gets more complicated from there. If we add an R to the DBMS we have a relational database management system. After reading several articles attempting to explain what this is, I feel I have the vaguest threads of understanding.  In my mind, a relational database management system allows someone to access a big pile of data in different ways depending on what they are looking for from one instance to another.  I may be completely wrong, but I think of someone with Amazon pulling out a list of everyone that bought baby supplies in the month of December to send one type of advertising and then a list of people who purchased expensive electronics to send another.  All from the same collection of sales data for the month. This is a pretty useful marketing tool and explains why at the bottom of the page, Amazon is offering me suggestions of things to buy based on past purchases.  Some of them look pretty good!




No comments:

Post a Comment