Return Home

Single Sourcing
By Suzanne Hardy

Author Bio

Single sourcing is an industry buzzword that actually has more than just "buzz" to it. In these lean times of one technical writer doing the work of three, the concept of single sourcing is more appealing than ever. Even if your company has enough writers to cover the workload, the idea of using one set of source files to create multiple outputs has definite allure.

Primarily, single sourcing can reduce the time spent maintaining multiple sets of files used for different deliverables. For example, if you maintain a large printed reference manual that contains information straight out of your help system, do you make changes in two sets of files? If so, not only does this method take plenty of time, but it increases the likelihood of typographical and formatting errors, not to mention missed information in one of the outputs. Single sourcing can eliminate these problems.

Single Sourcing Tools

So is it really possible to single source? The answer to this question depends partially on the tools you use.

For example, various features in popular publishing and authoring tools provide ways of hiding certain information in the output. Conditional text (or "build tags") is one such feature. This process involves "tagging" the content you want to exclude from a selected deliverable (such as a version of your online help that will be delivered with beta software).

Then, before you deliver the files, you select an option in the software to exclude content marked with the tag. In another tool, you can use fonts or templates to exclude information. There are also large content-management databases that store pieces of information that you assemble as needed, but these systems can be expensive and are typically used at companies with large, complex documentation sets.

Planning and Up-Front Work

If you decide to single source, a solid plan and some up-front work can ease the transition.

You will need to start with a single set of files that contains all the information you need to deliver (such as a help system or document files).

Then, analyze the content of the files to determine which information needs to go into each output. (If you have already been delivering the outputs, this process is straightforward because you know where the content belongs.)

Then, depending on the tool you are using, tag the information (or use another method) that will exclude the content from the output. For example, if you need to deliver a printed manual that includes content from your help system, you can tag any text that says, "Click here for more information," or other references specific to online help. Or maybe your manual is only a subset of the online help, in which case you could tag all information that is not going into the manual (including graphics).

Next, test the deliverables thoroughly to ensure that the right content is going into the right output.

Efficiency

Once you have analyzed, tagged, and tested the existing information, continue this process as you create content. When the system is in place, you will find that it’s more efficient than re-creating the information during the stress of deadlines, and it gives you more time to concentrate on your content.