After finishing “Coders at Work, “ more on that in a future blog post, and having little experience with non-RDBMS databases, I picked the book “Seven Databases in Seven Weeks” by Eric Redmond. The book appears to be of similar quality to it’s sibling “Seven Languages in Seven Weeks” by Bruce Tate.
The book starts out with the Postgres database. At the time of writing, this database wasn’t as popular as MySQL however it does make a good starting point as a baseline of comparison. It represents the “old guard” of databases. For most of the first week, I found that the first half of the first week was not of much interest to me. However, the fuzzy search extensions and full text search extensions caught my attention. I have always been aware that the capabilities existed, however, I never knew how they worked. Additionally the downloadable source code helped with creating a testing environment right out of the box. This was the same case for the “cube” extension/datatype. I found it very exciting to find out that you could do some rather interesting operations with multidimensional data and queries. I can’t claim that I’m an expert on using these features but its rather nice to have some hands on experience for it.
I don’t believe that having that content was the greatest value of the book. I believe what gives the book the greatest value is that investigating more on the cube package it led me to finding an online directory of the available extensions. I found the Postgres Extension Network. How exciting is it to find a directory of extensions to a fairly standard database that allows you to do some cool things? You can find extensions to interact with JSON data, store bitmaps, keep key/value data, additional aggregation functions, weight averages (This is a VERY interesting addition), and even attempts to do a “connected regions” logic within data items. These are reusable components that others have created, and that I found that I could get the database to perform these actions rather than code them myself.