What Is a Database?
Friday, January 4, 2008 0:43 by dChengOk. It doesn’t seem that hard a question to answer, but there are still a lot of people who don’t think of relational databases when they think of how to build applications in LongJump. Yet it is the most basic structure in computing and especially web computing. Consider the most popular online sites and services: eBay, Amazon, iTunes, Netflix, Facebook, MySpace, LinkedIn. At their very core, if you took away the whiz-bang features and the great content, you would have in essence relational databases.
So what does that mean?
Databases are these storage areas for information – essentially file cabinets converted to digital format. To store data, you have data records, which is data that complies with the structure you’ve defined.
In the case of iTunes, for example, the structure is around songs, artists, and albums (also called objects or tables in database lingo). Each object also has a record structure, for example: a song has a title, length, cost, date, etc. and also points to the album and artist. An artist has their name, description, and all the songs and albums their on. And each album has a name, price, and a collection of songs, and might also be related to a specific artist or artists. Everyone one of these objects might also be categorized as another way to dissect the data.
This is classic relational database structure.
Glue-Based Administration
What’s interesting is that the actual sum of a database’s parts constitutes a nebulous series of pointers or joins hooking one bit of data to another. It’s the glue, if you will. And in the old-school world, unless you’re a database administrator, you’d have no chance of ever changing that data structure. In fact, it often takes teams of people to hold the thing together, because one wrong move and your data can be lost or mangled beyond recognition.
Administrating this data is part of what LongJump has tried to simplify most. LongJump lets you define your structures however you need to and we try to make it bullet-proof for you. The result is that you get away from the management side and start doing stuff with the data, like analysis, running reports, filtering for information, or applying the data to team-processing, project workflow, or data tracking.
Looking Before You Leap (Jump)
But if you’re creating your own application, it always comes back to something even LongJump can’t do for you, which is defining what your data should look like and organizing it in a way that makes sense in a relational database world.
So before you start designing, start mapping out how you think your data should look. In the mean time, here are some nice resources for you to read up more on database theory:
