On the Design of Terminology
- have a common understanding of the ideas, concepts and other knowledge artifacts1 that are relevant for realizing these objectives,
- are a linguistic unity, i.e. they use the same terms to represent these ideas and concepts, and
- are able to provide themselves with assurances that everyone consistently and consequently uses the same term for the same concept.
These three elements form the basis of cooperation. They do not imply that the members of the community agree about everything, but it does mean that they have a solid foundation for formulating and discussing ideas, arguments, etc. They are a means to mitigate the risks (and associated costs) of misunderstandings, primarily by preventing such risks to materialize.
Note that preventing misunderstandings is not always a good idea. Many jokes, comedies etc. are based on misunderstandings occurring. Also, there are many situations in which the risks of misunderstandings aren't so high to warrant putting effort in preventing them, such as social talks at a coffee machine.
Our focus is on contexts in which misunderstandings do have consequences that are serious enough to warrant spending time, effort and resources in creating a common understanding, a linguistic unity, and providing the assurances that this is all actually the case.
TL;DR: In the subsequent sections we will:
- illustrate the importance of realizing a common understanding,
- elaborate on the difficulties that people encounter as they try to do so,
- propose a new approach that aims to overcome such difficulties and realize a provable common understanding.
The importance of 'linguistic unity'
Many cultures have stories, similar to that of the Tower of Babel, that observe that the big feats, such as building a "tower, whose top may reach unto heaven", cannot be achieved unless the people that do this have a linguistic unity, i.e. they use terms that they all understand in the same way. For example, if two railway companies have no linguistic unity, they may have different ideas associated with the term track gauge, which results in a risk that trains cannot run on one of the company's track.
The architecture of the building of the EU parliament in Strasbourg intentionally resembles the Tower of Babel as painted by Bruegel (see Figure 1), as a constant reminder a common understanding and a linguistic unity are required in order for them to (dis)agree.
Figure 1. The EU Parliament building resembles the Tower of Babel by Bruegel
The difficulty of realizing linguistic unity
Many people that have been part of project teams that engage in large projects have experienced this in person, and can tell many stories about how gaining a workable linguistic unity wasn't really achieved in some project, even though project members all agreed on how important this is, and real attempts were made to realize it.
The most common approach for this is to create a glossary, i.e. an alphabetically sorted list of terms and their descriptions (sometimes also called definitions), that the participants in a project all agree on. Usually, this means that people start to collect terms and descriptions that they either make up, or copy from other existing glossaries. Then, discussions follow that typically result in the modification of descriptions, as well as in more discussions. At some point, the effort needs to stop, and the glossary is established (published, finalized).
A glossary that has been created this way typically does not establish the linguistic unity that was attempted to be realized. If it had, you would not find documents (articles, blogs, mailing list messages, github issues, etc.) that used a term in a meaning that differs from the meaning as defined in the glossary they supposedly use. Also, you would not see discussions between people that each use a specific term in their own way, i.e. in a meaning that differs from that of the other persons.
We need some additional text here that on the one hand acknowledges the genuine efforts of people to achieve linguistic unity, and explains why these efforts so seldomly result in the intended effects.
Also, perhaps, some texts that comment on the quality (i.e.: usefulness) of definitions, e.g.:
- terms that are defined in terms of themselves;
- terms that are followed by a description of which it is unclear how that defines a term;
- terms that are specified as a criterion so that people can determine what is (not) an instance (example) of the term;
- terms that are specified such that, when a sentence mentions a term, and the term is replaced with its definition, the sentence is still grammatically correct and meaningful.
The approach called Terminology Design
text needs to be written here
- ideas, concepts and other knowledge artifacts are (intangible) parts of a knowledge. One could say that they constitute 'the mind' of parties. They need a (tangible) representation, e.g. terms, figures, sounds, etc. that parties can use to communicate with each other. The problem is that each participant of a community must be sufficiently sure that not only they all use the same terms, but also that they have a common understanding of the actual concept that an individual term represents.↩