Think of them as preformed queries.Macros and ModulesMacrosand moduleslet you add functionality to your database. Forexample, if you want to do a monthly cleanup of all records olderthan a certain date, you can automate that by creating a module. Foranything that you would have to run manually, you can probablycreate a macro to make the process faster.Non-relational DatabasesRelational databases are great for holding structured data, but notall data fits into a nice, clean structure with rows and columns. Twoother types of data that people need are nonstructured and semi-structured. These types of data can be housed in non-relationaldatabases. Before diving into the types of non-relational databases,it’s important to understand the key features of non-structured andsemi-structured data.Non-structured DataMost data in the world—80 percent by some professional estimates—is non-structured data. Examples of non-structured dataincludepictures, videos, web pages, emails, documents, texts, and socialmedia. Realize that this data has structure within itself. Forexample, emails have a sender and receiver, a subject line, and amessage body. The data is considered unstructured because itdoesn’t fit neatly into a database.Here’s a good analogy to help think about dealing with unstructureddata: Imagine you have a file folder on your computer. In thatfolder, you’ve placed photos from a vacation, emails, work history,515
school transcripts, text messages, all of your social media content,and X-ray images from a medical procedure you underwent a fewyears ago. That folder is now your “database,” and the content in itis your data. You can see that the data has no structure, but theremay be interesting links between different pieces of it. It will bequite challenging to find those links, though, considering all of thedifferent data types included.Non-structured data is often referred to asunstructureddata.Semi-structured DataAs its name implies, semi-structured data fits somewhere inbetween structured and nonstructured data. Specifically, semi-structured datais generally thought of as unstructured data thathas been tagged with metadata. Metadatais literally data aboutdata.Take the vacation photos and X-ray images from the unstructureddata folder, for example. Both of them are images, which are justcollections of pixels—how would they be included in a search? In asemi-structured world, tags of metadata could be associated withthe pictures to provide some context (maybe the date they weretaken, the location, or even some more specific information such asit included a hand). Granted, the hand picture could be either froma vacation or an X-ray, but at least you’ve got something to go on.Emails can be tagged too by listing the sender and the recipient,time, and date. Going back to the messy unstructured data folderexample, you can probably think of ways to include metadata foreach type of data mentioned.