Using metadata to research context – by Michelle Knight

Image Courtesy of UA Magazine

The image above shows why a technical writer needs to research context before creating any documentation. Should the author describe the picture as a vase or two people facing each other? Both assumptions would be correct. The answer depends on the type of document, why it is needed, and the intended reader. Metadata, background information about a resource (whether an image, another paper, or person) guides the author on fitting content to the circumstances.

Writers inform readers by using mental models that fit their audience's existing knowledge. Finding out this best framework to use can be overwhelming upon turning over each data asset and tribal practice. In my article "What technical writers need to know about data governance," I note that standardizing metadata through data governance makes it easier to find out about the data context. The writer can spend time on the most critical information references first.

This article picks up from the one on data governance, explaining what contextual information benefits writers, why use metadata, finding relevant metadata, and using that metadata to narrow in on the most important references.

What contextual information benefits technical writers?

Technical writers and any kind of researcher need to know contextual information containing the:

• Purpose
• Owner/author
• Timeliness

This background checks the trustworthiness, quality, and appropriateness of any cited works. 

We can see how contextual information works by running through an example to assess a website's validity. Go to the "About Us'' section on any website and find the organization's or creator's purpose, expertise, ratings, and recentness of the available content. For example, the non-profit Mayo Clinic passes as a good source for health information. We know from the professionals listed, operations, Copyright date, and testimonials from other authorities like U.S. News & World Report, on the "About Us'' page. We can be confident about the reliability of the Mayo website.

When we cannot find any information about the website's organization or creation, we need to question its information. Without the context, we would want to check another reference — a reputable one — to verify the website's truthfulness as it had no contextual information.

How did this translate into technical writing? Good technical writing requires good research and trust in the content, especially in ambiguous data. Without knowing the context of our citations, through the purpose, owner, and timeliness, our readers and we have less certainty about our document's reliability.

Why use metadata first to research context?

For three reasons, the technical writer will want to use metadata to research his or her resources' contexts:

• Format: We come across primary (first-hand) and secondary (reported) resources packaged differently. They can show up as images, audio, video, text messages, and chat with other people or machines. As technology continues to advance into the future, the number of information media will only continue to grow. Metadata makes it easier to consider all these types that can provide solid backing for written content.
• Volume: The amount of information has continued to grow exponentially, and the technical writer just does not have time to research every potential fact. Sure, you could meet with the document's owner/s or managers to get recommendations on resources, which would be an excellent first step. But you may miss other critical resources.

For example, you may need to meet with more than a project's business analysts and developers. You may need to meet with the company legal office or do some research on the internet to communicate data privacy and comply with recent data privacy laws. Failure to do so may mean you need to rewrite to avoid potential fines. 

• Speed: Information needed for a document can change fairly rapidly. What a subject matter expert or an article says one day could be outdated the next based on new information. Metadata describes the reference material's time frame since the last update or creation. This information tells the technical writer whether the information researched is most recent and prevents providing out-of-date content.

If technical writers need metadata to research the context, then where do they go?

Where do technical writers find metadata describing context?

Plenty of sources instruct writers to create metadata, categorizing their content in some management system. Advice to authors includes details on creating taxonomies, topic trees that list and structure terms and their relationships, and then using these to create metadata about their content. Often, the technical writer picks up the Darwin Information Typing Architecture (DITA) tips to better find his or her content and reuse it. This guidance recommends recording information about the content type, creation time, and status to remind you of your document's purpose and background. DITA tools help streamline metadata and optimize it for search engines or do SEO.

While metadata in a content management system helps understand the context, critical metadata exists beyond it. Contextual metadata can be found across an enterprise, covering machine-generated logs to departmental spreadsheets. Many colleagues from other departments may track contextual metadata manually. Technical writers should realize that they can find metadata about useful resources scattered across the organization.

Over 70% of enterprises have started to recognize the metadata's importance. They have begun to or have made plans to manage it in the future. As enterprise data governance initiatives mature, they will be better at organizing metadata. Plus, software tools exist to organize and catalog enterprise-wide data. In the meantime, a technical writer would do well to check in with the data governance initiative or the Chief Data Officer (CDO) to find contextual metadata about resources. 

When technical writers look for articles and books for background about the content, they will write, library database tools, and catalogs provide needed contextual information about resources. Also, bibliographies list metadata about different references. For example, Gale OneFile lists leading periodicals and journals. You could also search Google. However, the information retrieved may not necessarily specify the information's purpose, author, and recency.

How do you use the metadata you find to research context?

First, you need to clarify what kind of contextual information you want to find out about your information resources. Metadata about them exist in different formats, commonly three types:

• Business: Business metadata defines the general context of a company's contents, products, and conditions. The metadata relates to business purpose and strategy. For example, a business glossary lists terms or business metadata across an enterprise. Also, data governance practices provide business metadata to inform researchers about who owns the data, its purpose, and how frequently data gets refreshed.
• Technical: Technical metadata describes the logical and physical data context of the data platforms. This metadata type includes models of databases, access permissions, and backup procedures, among other guides.
• Operational: Operational metadata specifies the situation around processing and accessing data. Maintenance plans, logs, and job execution comprise examples of operational metadata.

If you search for secondary sources, like articles or webpages, you will find bibliographic metadata about the publication, title, location, format, author, and subject, among others. This type of metadata makes certain content more findable and connects with the DITA metadata elements mentioned above. You add bibliographic metadata, which can overlap with business metadata, to your documents.

Knowing how to use metadata to research context requires planning and understanding what kinds of questions about the purpose, authors, audience, and publications you need to ask. Once you have the information you want to find, make a plan of attack for searching the resources. Then you can organize your notes into a mind map and use your DITA metadata elements to refresh yourself about the context of that information.

For example, say you had to write a white paper about data management from a survey of about 200 participants. You would want to find business metadata to understand the survey's purpose, owners, and caretakers during the survey's execution and processing of its results. You would need to have technical metadata about what questions the survey tool will cover, its strengths and weaknesses, and the login information— to understand how the users would access the survey. You would need operational metadata about who to consult for analysis of the responses, who has written similar documents and their processes, and a style guide to organize the documents and proceed with the writing. 

After knowing what metadata you need to find, you can then list the sources you need for research: the survey website, past papers for a similar audience, business glossaries, managers to consult, and interdepartmental meetings that require attendance. From there, you can organize the sources you need to consult and the importance of the information you obtain to the context needed for the white paper.

Conclusion

Technical writers have to do adequate research before creating technical documentation to generate trustworthy quality content. They need to research context to know how to frame the content and learn what information is correct, especially when it is ambiguous. Using metadata to research context saves time to get the most relevant information fit for a document's purpose. As organizations expand their enterprise data governance programs, they can point to critical metadata located in other departments' information resources.