Creating a game-design ontology (why do it?) 3/*

Reading a thread on Bluesky by Sergi Valverde about the evolution of software, I thought it might be fruitful to draw parallels with the evolution of game design for tabletop role-playing games.

How so?

First of all, the study of software evolution is quite convincing. They are often viewed as living systems traversed and guided by evolutionary forces.

« we see the same evolutionary forces: imitation, recombination, competition, parasitism, symbiosis. » in Sergi Valverde, Ricard V. Solé; Punctuated equilibrium in the large-scale evolution of programming languages. Journal of the Royal Society Interface 2015; 12 (107): 20150249. https://doi.org/10.1098/rsif.2015.0249

Far be it from me to claim mastery of the concepts, much less the basics, of complex systems analysis with an evolutionary angle, but I’m launching the reflection. However, I believe studying role-playing games could greatly benefit from methodological discoveries in other domains and, cautiously or boldly, import them.

Usually, in the study of complex systems, there are graphs with relations between many elements and also many variables (more or less measurable) to understand the interactions between elements.

Parallels between software and TRPGs

Both cultures share many similarities:

• Emergence and growth in the same parts of the world, in close generations, even with a similar sociology of their actors.

• Design of cultural products based on rules, procedures, recipes, ways of doing.

• Co-existence of a proprietary market and a “free or open” market.

• Inspirations, influences, movements, practical philosophies for each product.

• Access to the full corpus of games (their DNA), even possibly full indexing to obtain rich metadata,  just as code is for software.

• The same evolutionary forces mentioned above.

Here, another fascinating study by Valverde et al. studies the evolution of arcade video games.

Valverde S, Vidiella B, Spiridonov A, Bentley RA. The cultural macroevolution of arcade video games: innovation, collaboration, and collapse. Evolutionary Human Sciences. 2025;7:e30. doi:10.1017/ehs.2025.10015 

They find trends present in biology: “diversification, collapse, persistence.” According to the authors, technological objects offer a richness of description easier to obtain than biology and thus allow testing evolutionary theories. For example, as in nature, computing power and ROM memory of arcade games create constraints. Game clones can experience stagnation while collaborations between game publishers can create innovation.

What should be done

To properly analyze these evolutionary trends, one should:

• design an ontology (an organization of information) describing games and allowing fine-grained indexing of game mechanics.

• index games according to this ontology.

• do it for a representative sample of games (to be constituted), perhaps weighting this sample according to the notoriety of the game.

• process the results using mathematical methods I don’t master.

• (possibly) draw conclusions.

• the hardest part: absorb the literature on the evolution of cultural objects.

  • Understand the concepts.

  • Verify which ones could be adapted to tabletop role-playing games.

  • Create new ones?

What has already been done

• I started to design an ontology to describe tabletop role-playing games. It is described in the two previous posts.

• I also created a prototype in 2019 in the form of a Timeline Tree of Tabletop Role-Playing Games.

• RPGGeek and LeGrog each contain a database of games properly catalogued — though with less developed indexing of game mechanics.

• I tried to write a Python script to parse RPG rule PDFs to detect what types of dice are used.

• I contributed to Wikidata by adding some role-playing games with more or less rich metadata. At present, I have mainly indexed the citations of each.

Future paths

• Outside the reach of course: because I don’t have the reputation or the platform of influence to launch it — it would require carrying out indexing of the entire corpus by hand, with an army of volunteer indexers.

• Apply automatic processing (as for dice) to other transversal elements of role-playing games:

  • mention of a screen (GM screen), a character sheet, stats, skills, etc.

  • presence of tables in the rules, or ratio letters/numbers.

  • presence of exotic names.

  • etc.

In my opinion, these elements can benefit from a valid automatic numerical indexing. Maybe one day, language models (LLMs) will be able to index advanced game design.