tdog.blog

What is database publishing? Is it the same as content management? File management? XML-aware publishing?

The short answer is “pretty much.”

Database publishing, as defined by Adobe, is this:

A process for managing, creating and publishing content through extensive use of database systems and content creation tools.

The basic idea behind database publishing is to store your data (price lists, articles, technical chapters, data sheets, or simple variables like name, address, occupation, etc) in a central repository, then push that data out onto a series of templates.

Some examples include:

• Reporting. Take a bunch of stored data fields and make sense of them to your audience. Opportunities for customized or personalized reports abound. For example, your VP of marketing may only want the top level numbers, but a marketing specialist figuring out better ways to handle event marketing would need more detail.
• Web site management. A CMS is simply an online database hooked up to an online templating system, with some kind of online management interface in between. It is database publishing through and through.
• On-demand printing. Lately there’s been a lot of talk about on-demand printing, particularly when combined with variable printing. Some examples:
  • Manufacturer with hundreds of products puts technical documentation into a central database, then allows users to visit their site and select which documents they want to have printed, either at a 24-hour printer or directly at their desktop. A more complex solution might involve grabbing only those sections they want and creating custom PDFs that can be printed at the user’s desktop, or combining different sections from different documents.
  • University system collects data on prospective students online, then takes this data to send them a fully customized prospectus. If the prospective student has an interest in music, physics and mathematics, emphasize this information and not the football games, poetry readings, and annual Greek Week.
  • Government agency collects information on various topics of interest to citizens, then categorizes this information. Citizens can log on and find information they need (such as phone numbers for a council member) or documents (such as the right form to fill out in order to register a DBA) or resources (the right place to call if you have sales tax questions), compile it into a custom PDF, and save it to their desktop. That way they can print this information out and keep it in a binder at their home or business.
  • Consumer-oriented company does all their business via direct marketing. They collect information on prospective clients (age, ethnicity, educational background, name, address, last 5 major purchases, etc) and prepares mailings using a straightforward cluster analysis. If, for example, they are sending out gym membership trial offers via postcard, there’s no need to send everyone the same post card with a big bodybuilder on the front. Perhaps professional women in their 30s would respond better if the image showed a fit young woman on a treadmill. Or maybe another mailing may improve if the company could co-brand the direct mail piece with the logo of the school the recipient attended. And let’s not forget that putting the recipient’s name front and center can increase response.
  • Wholesaling giant needs to track prices for all their SKUs on all its sales channels–retail operations, partner companies in Canada and Europe, direct mail catalog sales and more. Each SKU with price and other information is centrally stored in a database. A set of rules is applied to update all the channels on price changes and other edits made in the central database.

If most of this sounds similar to content management, it should. All we’re talking about here is the ability to centralize your data storage, then doing custom publishing of this data to one or more output devices or locations.

And yes, you can do much the same kind of stuff with XML. We’ve been particularly impressed with the Berkely XML DB, which can store XML snippets as entire objects (no need to rip them apart and store them in relational rows and columns), which makes for faster retrieval.