It’s tempting to predict that advances in AI will alleviate the task of taxonomy design and the meticulous planning behind content terminology, the crucial task of identifying, storing, and managing data and content. After all, DeepL, with its $2 billion backing and the neurally networked Google Translate with its 1 billion installations have all but conquered language barriers in business, seamlessly integrating into team tools like Slack.
Surely AI will somehow automate the detailed hierarchy work behind taxonomy? Well, if terminology and language were fixed and universal, there may be hope. But it’s evolving, morphing, expanding, and developing as fast as AI itself, requiring sometimes subtle and sometimes quantum shifts in the communication patois.
In their post on terminology management, TERMINOLOGY Management: The Key to Precise Communication, PTI Global state: “Language is changing frequently, be it through generational changes, technological developments, or simply that a language catches up with new concepts by defining relevant terms.”
A taxonomy requires a disciplined scheme of terminology classification with a vocabulary that identifies categories and subcategories into a hierarchy with parent and child tags; it’s the backbone of a coherent content management system.
Yet, as digital communicators will know only too well, different industries organise and categorise themselves differently. A technical writing project is like a construction project: scaffolding is needed to keep control of its many facets. Taxonomy is the scaffolding when the technical writer builds it, and it disappears when the output is published to an audience.
Luca Vettor, in his Medium Post, The Backbone of Technical Writing is Taxonomy, argues that technical communication is knowledge management, which organizes information for a defined purpose, and that’s where taxonomies come into play. Firehead has loved knowledge management and classification before knowledge management and classification were cool. Thank you, Luca!
Ultimate goals for each taxonomy include:
- Fast and reliable content search and retrieval
- Improved content analysis
- Better management, governance, and compliance
- Higher categorisation accuracy
- Flexible reusability of content
As far as possible, using plain language is the key when preparing user manuals around technical terms and findability as search engines tend to be inadequate for specialised search repositories (which we argue good keywording would really help progress this state of affairs).
Using plain language reduces translation error, enhances the customer interaction and coherent messaging, and democratises the communication between customers and stakeholders: everyone is on the same page. For this reason, content users and stakeholders are often involved in the taxonomy design process, but their preferences vary from sector to sector and product or topic-specific contexts.
In a recent (May 2024) question and answer article about taxonomies, ontologies, and semantic layers on Progess.com, Jim Morris argues ‘The need for smarter approaches to data structuring and interpretation has never been more pressing.’
The enormous amount of data being collected by research and marketing departments needs further grouping by characteristics, attributes, and targeted relationships, whether broad-based or granular, again requiring custom-built taxonomies. The categories and subcategories may be archived on dozens of levels below the surface with very industry-specific inter-relativities.
With so much data now web-based, ontologies are relied upon for sharable and reusable domain knowledge with knowledge graphs modelling their complex relationships. Knowledge graphs give data semantic layers, building a framework for data integration, unification, analysis, and sharing.
‘In recent years, there has been an uptake of expressing ontologies using the Web Ontology Language (OWL), a semantic web computational logic-based language, designed to represent rich and complex knowledge about things and the relations between them.’ You can read more about OWL and its applications here.
The lack of ‘global’ and even industry sector standardisation of taxonomies appears to be a contradiction in terms when considered against their purpose. This lack of standardisation means we are years, perhaps decades, from being able to expect AI to mine industry-specific taxonomies without huge discrepancies in its search results and generative output. Overlay that with the rules, axioms, classes, and attributes that drive domain-driven ontologies, and the challenge becomes even more daunting.
But we are wise enough to know every cloud has a silver lining. A whole new, exciting and dynamic industry is emerging within digital communication. It’s one that will require a huge cohort of language specialists grappling with what must become a complex global ‘Esperanto’ of taxonomy, ontology, and semantic layers if AI is to integrate itself within them and ‘neurally network’ content and data organisation.
And you can be part of that if you visualise the bigger terminology management picture and train yourself in the relevant disciplines!
If you wish to further your career in the fast-moving, ever-changing world of technical communications, sign up to the Firehead Training Academy today.
