We are torn what to think of glossaries: on the one hand, everyone on the team is grateful when there’s a well-maintained glossary. On the other hand, no one wants to create and maintain it. The coordination process can also lead to disgruntlement and eat up a lot of time. And even if there is a good glossary: There’s a big risk that many stakeholders won’t hear about it and it will gather dust in a drawer.
We just released several glossary-related features of Semiant. Hopefully, we can help the glossary achieve a renaissance. We invite all readers to try out the latest version of Semiant.
Usefulness of a glossary
For a glossary to be useful, it must be accurate and used. When that’s the case, a glossary has a lot of benefits:
- Fewer misunderstandings save time, avoid mistakes, and reduce friction
- New employees can familiarize themselves more quickly and take up less of the experts’ time
- It is less error-prone to deal with product lines and variants
- Maintenance over the product lifecycle is simplified
Standard glossary or controlled vocabulary?
Why reinvent the wheel? After all, there are ready-made glossaries on many specialized topics, from “automotive” to “zoology”. And indeed, if there is a suitable glossary, we should use it. For example, INCOSE’s Systems Engineering Handbook includes a glossary (Appendix C: Terms and Definitions).
Standard glossaries are useful in education. But in projects, a standard glossary rarely covers all relevant terms. In addition, terms that are familiar to certain stakeholders may have different meanings. Therefore, controlled vocabulary is an important category of a glossary that can complement or replace a standard glossary. Ideally, then, we have both a standard glossary and a specific glossary. This can be project- department- or organization-specific.
Merge glossaries intelligently with Semiant with the browser extension
Semiant is a virtual quality assistant that performs mundane tasks in product development. To do this, Semiant hooks the glossary into any web-based application via a web browser extension (currently only for Chrome and Edge).
The current version of Semiant is minimal, but already supports multiple sources for the glossary and allows merging in the web browser. How this works can be seen in this video:
Development partner wanted
As you can see in the video, Semiant is currently a minimal demonstrator. However, this shows that the technology works in principle. We have many questions and ideas on where to go from here. We are looking for a development partner who recognizes the potential and wants to help determine the further direction os Semiant. Specific ideas and questions are:
- Which data sources should we support? Currently Semiant reads existing glossaries from RDF/Turtle data sources.
- How useful is glossary term extraction via Natural Language Processing (NLP)?
- What team and categorization capabilities should we implement? Also, how expressive should the data structures be?
- How important is the use case of gathering contextually relevant information from applicable documents and standards?
- How can we improve the usability of the glossary with the browser extension?