A Metascience Study of the Adoption of Low-Code terminology in Modeling Publications

Low-code is trending and replacing other similar model-based practices, such as Model-Driven Architecture (MDA), Model-Driven Development (MDD), Model-Driven Engineering (MDE),… At least in terms of the terminology used by many tools.

Despite the potential technical differences between the two concepts, both Low-Code and traditional modeling share commonalities and the goal to accelerate the creation of software systems. But beyond the technical aspects,

• Can we observe differences in the communities backing each paradigm?
• How this novel Low-Code area is consolidating itself in academia, in particular within scientific conferences covering software modeling?

Such analysis could benefit both communities, especially to help them better understand their relationship and, therefore, their potential synergies and growth opportunities.

To perform the analysis described above, we focus on answering the following three questions:

1. How the number of publications related to Low-Code and traditional Modeling have evolved over time?
2. Who are the authors of such publications and do they overlap, i.e. publish both low-code and modeling publications?
3. What are the most common publication type and venue for Low-Code publications?

 

Metascience analysis of low-code publications

To answer such questions, we used Lens.org to collect articles containing “Low-Code” or “Lowcode” in their title, abstract, or keywords; and articles containing either “Model-Driven Architecture”, “Model-Driven Development”, “Model-Driven Engineering”, “Model-Based Architecture”, or “Model-Driven Software Engineering” in the same fields. As a result, we collected 1.160 articles related to Low-Code; and 12.840 articles related to traditional Modeling. Then, we manually cleaned the Low-Code set, removing 435 articles from it as those used the term “Low-Code” in a different context, were not made publicly available, or were duplicates. Due to the impracticality of manually cleaning larger collections, we did not perform the cleaning step in the set of traditional Modeling papers. At last, we performed a series of analysis, with their results illustrated and discussed below.

1. Evolution of Low-Code Publications

In Figure 1, we see that the first Low-Code publications appeared in 2017 and its presence in the literature has been increasing since then with a steep rise. Additionally, we note that in 2023, the number of Low-Code papers surpassed the number of “traditional” Modeling papers. On the other hand, in Figure 2 we see that the number of papers mentioning both paradigms is still negligible.

 

2. Authorship Analysis

In Figure 3, we observe that the overlap of authors who published in both traditional Modeling and Low-Code is small, around 16% of the analyzed authors. This percentage increases to 28% when analyzing only the most influential active authors of each area. Furthermore, 10 of the most influential and productive authors in Low-Code are also equally influential in the traditional Modeling community. On the other hand, there are also authors who are highly productive in Low-Code but have no publications in traditional Modeling.

3. Low-Code Publication Profiling

In Figure 4, we see that Low-Code publications are mostly published in workshop and conference proceedings, with 48% of the articles being published in this type of venue. In Figure 5, we observe that the most popular venue for Low-Code articles is the companion proceedings for the ACM/IEEE International Conference on Model-Driven Engineering Languages and Systems (MODELS-C), publishing workshop, tool and poster papers of the conference, with almost five times more Low-Code papers than other formal publication venues.  Note that, in part, this number is influenced by the papers of the “LowCode” Workshop, collocated with the MODELS conference since 2020.

Some initial reflections

•  Low-Code interest is helping the Modeling community: In five years, the yearly number of Low-Code papers has reached a value higher than the yearly sum of all Modeling papers, which has helped maintain the overall number of publications in the field and even grow it a little bit. It’s still early to conclude whether we are just seeing a migration pattern (the same authors that were publishing about modeling do the same but calling it low-code instead) or we are actually bringing new “blood” to the community. This seems to be the case at the workshop level but it remains to be seen how this will translate to more mature publications.
• Workshops and arXiv as entry points for Low-Code: Our results reveal that Low-Code is entering the Modeling community via workshops, mainly from the MODELS conference. We believe that this behavior aligns with typical scientific publication procedures, where novel ideas are typically first published in “entry-level” venues such as workshops, posters,…  but it requires further validation. On the other hand, the high presence of publications on arXiv may demonstrate new trends in publishing research work. Consequently, we anticipate more Low-Code publications being published at main conferences and journals, over the following years as the current workshop publications (and the authors behind them) mature.
• Do not reinvent the wheel: The fact that there is a 28% of overlap between the most influential authors in Low-Code and Modeling may reveal that senior authors from the Modeling community are adopting the Low-Code terminology and techniques, and, consequently, influencing younger co-authors. For this reason, we emphasize the importance of senior researchers teaching the younger members of the community about the need to know prior modeling work to avoid reinventing the wheel
• Open community and democratization of technology: Low-Code platforms are often used to produce software in diverse areas apart from Computer Science such as Medicine, Meteorology, Business, among many others. Accordingly, we observe that publications related to these software applications are commonly the first experience of authors with modeling-related publications. This could be an opportunity to expand the modeling/low-code techniques beyond our core field.

You can read an expanded version of these analysis and reflections in our arxiv paper. And let us know whether you agree with our reflections and/or what other analysis you’d like to see!