This post continues the topic of creating a robust data dictionary in PostgreSQL. My previous post on building a data dictionary in PostgreSQL focused on using built-in psql tools, such as \dn and \dt. It also quickly covered the use of comments to document database objects. Those tools and shortcuts are great and readily available, but beyond quick exploration tasks these built-in options don't provide enough functionality. This is especially true in today's data-filled world with increasing regulation.

This post is part of the series PostgreSQL: From Idea to Database.

Meet pgdd: PostgreSQL data dictionary

The PostgreSQL Data Dictionary (pgdd) project, available on GitHub, intends to improve our ability to document databases in PostgreSQL. This open-source project super-powers your PostgreSQL databases by introducing additional meta-tracking abilities. It also makes the internal database meta-data more user accessible and user friendly through a handy set of views. These views can be queried and joined using standard SQL syntax using any tool you like.

pgdd exists to make PostgreSQL data dictionaries more usable.

Data dictionary should be for everyone

Imagine an analyst with access to a PostgreSQL/PostGIS database. That database probably has data from a variety of open and closed sources. That analyst spends quite a bit of time in QGIS, but very little time in psql. Being able to run standard SQL type queries in their favorite tool makes the data dictionary more usable for this user.

Continue Reading

This post examines the work of producing analytics/visualizations and the role software plays. Business Intelligence (BI) software, such as Power BI, Tableau, and SAP Dashboards are a few examples of software with a lot of buzz. The marketing folks in that industry would have us believe their software enables analysts to point and click at data and enable magic. This magic feature provides instant access to interactive dashboards, drill-down functionality, and more. Don't forget, the data will be reliable, accurate, easy to share, easy to secure, and cross-platform (mobile-friendly) too!

Reality check

Data visualization software in 2018 doesn't have the magic feature working yet. I believe it takes at least as much human effort today to produce a great visualization as it did a decade ago. The improvements in the software are not making the process quicker or easier. The reality is that working with data isn't for everyone, and #DataIsHard.

I recently attended a PASS webinar by Scot Reagin who stated:

"90% of users are not self-service capable."

Continue Reading

This post provides an introduction to building a data dictionary directly in our PostgreSQL databases. The steps outlined here are specific to PostgreSQL, though every database platform has these basics components in place.

This post is part of the series PostgreSQL: From Idea to Database.

What is a Data Dictionary?

Documentation is a good thing. Data dictionaries are one important component of documenting your databases. A data dictionary is a common tool used to provide database-specific documentation to analysts, developers, and other business users. A good data dictionary provides insights into a database's structure, constraints, relationships, and sources of data found inside a system.

Continue Reading

In general, the core PostgreSQL community doesn't latch onto GUI tools. That said, pgAdmin is one of the more popular GUI options for PostgreSQL. I hadn't used pgAdmin for quite a while (the transition from III to 4 was... rough), but a recent feature caught my eye: Geometry viewer.

This feature, available in pgAdmin 4 v3.3, allows you to see your PostGIS data in a separate tab directly in pgAdmin. When you execute a query that includes PostGIS data, pgAdmin overlays your data on the OpenStreetMap background in the geometry viewer.

Awesome!!

Boston GIS introduced the basics of how the Geometry Viewer functions. This post illustrates why I think this feature is so cool.

Why is geometry viewer awesome?

Spatial (PostGIS) data typically doesn't translate well in most SQL tools. pgAdmin's new geometry viewer provides a dead-simple way to visualize, verify and troubleshoot spatial data in their databases. This feature is targeted for database professionals who support PostGIS databases.

Continue Reading

The database is one of the most critical components of most software projects. Planning for your database project should not be taken lightly. Failure rates in IT projects are unfortunately high; much of that problem is due to a lack of proper planning and design. Most of the database anti-patterns I previously wrote about can be avoided by understanding the system you need to build and designing it appropriately.

This post uses the PiWS project to illustrate the Define and Design approach I use throughout the initial development phases of any new database project. Most projects go through multiple iterations of define and design.

This post is part of the series PostgreSQL: From Idea to Database. This series uses the PiWS project for our example project, to read more about the PiWS, see my introductory post.

Project management, not optional

Anyone who has worked on a project with more than one person to design and build anything has heard this: "I forgot to tell you..." followed by some various change requirements.

Continue Reading