CIDM

Spring 2025

The Fabric of Our Lives

Headshot of smiling women with short straight hair wearing glasses Dawn Stevens, Comtech Services

 

 

 

At DITA Europe in February, I attended a panel talk from the Component Content Alliance on Data Fabric. The panelists discussed the parallels between DITA and data fabric technologies, exploring how structured content standards can enhance user experiences and support robust business analysis. Yet, even as their abstract touted that the integration of DITA with data fabric technologies could “unlock new possibilities for data connectivity and analysis in an increasingly complex digital landscape,” the panelists themselves admitted that the terminology — Data Fabric — might be new, but the goal – connecting disparate data systems – is not. One panelist even suggested that new terminology is necessary to keep interest in the subject piqued – an occasional rebranding of sorts until one or more approaches to achieving complete connectivity are fully realized.

To be fair, the term is not that new; my brief exploration of the subject indicates it’s been around for a couple of decades in the data warehousing space, referring to a “data management and integration architecture that allows organizations to access, manage, and share data across multiple systems and applications.” Its use in the context of structured content is more recent, but reflects the same goal: an ability to integrate diverse content sources into a cohesive, accessible, and intelligent system.

As I listened to the panel and contemplated the term “fabric” within the context of our longstanding challenges in providing ready access to all relevant content, regardless of source or format, I found the word really resonated with me. I am a member of an aging population that was required to take home economics in school. In ninth grade, one period a day was spent learning to cook and sew, and in the decades since then I’ve used those acquired sewing skills intermittently to join all sorts of fabrics into a variety of end products: blankets, clothing for me and my family, Halloween and theater costumes, and even a few bridesmaid dresses. The final results of my efforts, coupled with the occasional challenges in getting to the final product, lead me to the conclusion that the word “fabric” is an excellent analogy of the issue at hand, highlighting the vision and the challenges of the goal.

The Vision

The vision of a data fabric, in the context of our world of content, is to transform isolated content units into a unified resource that supports informed decision-making, innovative analytics, and efficient operations. Cloth fabric is produced by weaving or knitting various fibers together into a comprehensive whole—a finished product that is stronger and more versatile than its individual component threads. The individual strands are no longer meant to be distinct but part of a utilitarian whole. Each thread contributes to the overall look and strength of the end product, but the end user views the end product as a whole; that is, while I may know that a fabric comprises thousands of individual threads, the bolt of cloth, not its individual threads, is what I work with to reach my end goal.

This process of creating cloth helps to illustrate the required process for creating a data fabric:

  • Individual Threads as Content Sources. Consider each data source—be it technical publications, marketing, training, or service documentation —as a thread. Just as threads come in different colors, textures, and lengths, content originates in diverse formats and from various systems. Each thread on its own is useful, but its full potential is only realized when it is part of a larger, coherent design.
  • Patterns as Content Models. Before sewing, a tailor selects or designs a pattern that dictates how the threads should be arranged to create a cohesive fabric. In data fabric, this pattern is akin to content models and schemas that structure the content. These models ensure that the content is organized, easily interpretable, and aligned with the organization’s goals.
  • Stitches as Metadata. The act of sewing involves stitching threads together, binding them into a single piece of fabric. Similarly, metadata “stitches” disparate content sets together, providing relationships between the various pieces. This integration creates a robust framework where data flows seamlessly across systems, much like how stitches hold a fabric together, providing durability and coherence.
  • Needles and Sewing Machines as Content Integration Processes. A skilled tailor uses both machine and handwork to precisely stitch pieces together. The sewing machine ensures efficiency and durability, while hand sewing with a needle may be necessary for delicate work and to achieve specific precision. In a data fabric, machine learning algorithms and automation streamline data integration, detect patterns, and ensure real-time availability of information, but a human in the loop is necessary for delicate and precise decisions in the integration.

The Challenges

Although the analogy seems to work well at the micro, thread-level, I question whether our content sources are “threads” or are in actuality different pieces of cloth. Rather than making a new textile composed of differing threads, are we actually making a quilt, composed of radically different squares of material?

In sewing, you can’t simply stitch two textiles together and expect to get a functional garment. There are certain materials that are difficult to (or in fact shouldn’t be) sewn together in the same article of clothing:

  • The materials have incompatible washing instructions.
  • They stretch and shrink at variable rates.
  • They have disparate wear and tear tolerances.
  • They may require unique stitch types in the seams.
  • They drape differently on the body.

Similarly, you can’t simply combine many data sources and get a functional artifact either:

  • The sources have incompatible goals and objectives.
  • Their structures vary.
  • They have disparate shelf lives.
  • They use unique styles, voices, and tones.
  • They address different audiences.

It seems to me that the underlying reason that companies fail in creating an integrated content system is that they are sewing a quilt instead of weaving a fabric. They try to preserve each fabric square so that it retains its unique beauty, rather than tearing it down into its individual composite threads and reweaving it all together again.

Honestly, this approach might create something lovely and functional, just not as versatile as might be desired. A quilt makes a lovely wall hanging or a bedspread, but most people aren’t going to go out dancing draped in one. On the other hand, if we work to preserve each fabric square separately, we might ultimately find that some fabric squares are so different that they should never be directly stitched together or the end product would be undesirable – unable to be washed, too delicate or awkward to be worn, or frankly, just so ugly we never wear it. No one actually wants what we’ve created.

So the compromises begin. Tailors create complementary pieces—accessories that can be worn together, but also stand on their own. Individuals can buy only the pieces they want. And while these items may be displayed together in a sales campaign for a while, ultimately they are moved into different departments in the store. Buyers bring them home and they are then cared for and stored separately: our woolens in the cedar chest, our delicate laces in acid-free tissue paper. The best laid plans to bring the fabrics together are forgotten, perhaps occasionally being rediscovered that they could be worn together on certain occasions.

So seems the path to integrated content. We start in earnest to bring together our random assortment of content into some sort of pattern, but we are ultimately unwilling to break them down further into their component pieces and we create something no one actually wants. Users then pick and choose the pieces they need, individual departments reclaim control over those pieces, and the process begins again, under a new name.

Rather than repeat that cycle, let’s take a lesson from the current name, abandon our cloth remnants, and commission a master tailor to craft a new fabric, with every thread meticulously selected and every stitch thoughtfully placed into a beautiful, functional masterpiece that drives business success.

 

About the Author:

Dawn Stevens is CIDM’s Director and President of Comtech Services. She has over 30 years of practical experience in virtually every role within a documentation and training department including project management, instructional design, writing, editing, and multimedia programming.