Small Yet Configurable: Unveiling Null Variability in Software Conference: Preprint; ; CCS: Software and its engineering CCS: Software and its engineering Software product lines CCS: Software and its engineering Software configuration management
Xhevahire Tërnava, Georges Aaron Randrianaina, Luc Lesoil, Mathieu Acher · 2026
Read Small Yet Configurable: Unveiling Null Variability in Software Conference: Preprint; ; CCS: Software and its engineering CCS: Software and its engineering Software product lines CCS: Software and its engineering Software configuration management by Xhevahire Tërnava, Georges Aaron Randrianaina, Luc Lesoil, Mathieu Acher (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2604-15957.txt.
Introduction
Tërnava and three colleagues present the first empirical study of configurability in small-scale software, programs with limited code or binary size. Analyzing 108 GNU coreutils programs, they find even small programs vary widely, with up to 76 options per program. Run-time variability correlates strongly with codebase size, and across 85 releases of the 20 smallest programs, variability grows over time, mainly from added compile-time options. They coin null-variable software for systems with no configurability beyond mandatory features, arguing that trimming needless variability yields lighter, more maintainable programs.
Catalog record
| CREATOR | Xhevahire Tërnava, Georges Aaron Randrianaina, Luc Lesoil, Mathieu Acher |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 59228 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2604.15957
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Small Yet Configurable: Unveiling Null Variability in Software Conference: Preprint; ; CCS: Software and its engineering CCS: Software and its engineering Software product lines CCS: Software and its engineering Software configuration management Xhevahire Tërnava, Georges Aaron Randrianaina, Luc Lesoil, Mathieu Acher Abstract . Many small-scale software systems, that is, with limited codebase or binary size, are widely used in everyday tasks, yet their configurability remains largely unexplored. At the same time, studies on modern software systems show a trend toward increasing configurability, alongside growing interest in building immutable, specialized, and reproducible software. In this paper, we present the first empirical study on the extent of configurability in small-scale software systems. By analyzing 108 programs from GNU coreutils , we show that even small programs can exhibit significant compile-time and run-time variability, with up to 76 options per program. Then, there is a high correlation ( 0.78 0.78 ) between run-time variability and codebase size. Furthermore, an analysis of the 20 smallest programs across 85 releases reveals that variability tends to increase over time, primarily due to the added compile-time variability. This suggests that shifting options between run-time and compile-time, removing unnecessary run-time variability, or resolving compile-time variability early, can help reduce codebase complexity and size. We also introduce, for the first time, the concept of null-variable software system, one with no configurability beyond mandatory features. Our findings show that high configurability is not exclusive to large-scale systems and that reducing unnecessary variability can lead to lightweight, smaller, and more maintainable software. We hope this effort contributes to designing new software by understanding how to balance its configurability with codebase size. 1. Introduction Since the software crisis in 1968 ( 33 ) , significant advances have been reached in the development of variability-intensive software systems. Nowadays, software systems are highly configurable and expose to users their abundant configuration options. These options make possible to customize a software system for usage in different contexts ( 10 ) , without the need to modify its source code. At the code level, most of the options are set at compile-time or run-time ( 10 ; 9 ) , for example, as command-line parameters in C-based systems. Moreover, variability is often recognized as a strength of a software as users have the flexibility to adapt it to their needs. However, software systems are becoming more and more configurable, such as Linux kernel with more than 20 thousand of options and a configuration space ( i.e. , possible variants of Linux) of more than atoms in the universe ( 37 ; 32 ; 25 ; 3 ) . This ultra-large scale of variability is exceeding the human and even machine limits to deal with it, for example, to test and maintain it ( 17 ; 1 ; 51 ) . The ultimate and reasonable goal of researchers and developers today is how to address, manage, evolve, and reduce this large and increasing software variability. To our knowledge, the extent of variability and configurability in small-scale software is seldom analyzed, and the notion of a completely invariant system remains undefined. Actually, motivated by this gap, we asked several questions. What is the extent of variability in small-scale software systems, compared to the extensive configurability of large-scale systems? Is configurability also an important characteristic of small-scale software? Do software systems exist that exhibit little to no noticeable variability? There are studies that have used small-scale software, simplified versions of larger systems, or excerpts from their feature models, for various purposes, such as evaluating variability implementation approaches ( 6 ; 31 ) or debloating software systems ( 8 ) . But, in our knowledge, there is no work that explores such questions. In addition, the field of software engineering has traditionally focused on configurability and adaptability, leaving systems with no variability largely unexplored. While terms like "static", "immutable", or "constant" software are occasionally used 1 1 1 https://sonalake.com/latest/the-future-is-minimal-and-immutable-a-new-generation-of-operating-systems/ , there is no formal concept to describe systems designed to lack configurability, representing the antithesis of configurable systems. Herein, we empirically investigate the extent of variability in small-scale software by analyzing 108 programs from coreutils 2 2 2 GNU core utilities: https://www.gnu.org/software/coreutils/ and 85 releases across 20 of its programs spanning the years 2003 to 2022, including programs such as ls , cat , and mkdir . Most of these programs have implementation histories dating back to 1979 ( 18 ) . Our findings show that even small-scale software programs exhibit significant variability, with up to 76 configuration options and an average configuration space of 2^{15} . Moreover, we observed that run-time and compile-time variability have distinc roles in the dynamics of software systems. While runtime variability impacts the size of a program’s codebase at a specific point in time, compile-time variability appears to drive the growth of the codebase over time. Additionally, we identify two strategies for reducing codebase size, both centered on removing unnecessary runtime variability. The main contributions of this paper are as follows: • We present the first empirical study on the extent of variability in small-scale software, unlike previous studies that focus on large, complex configurable software systems or explore small-scale software for other purposes. • To the best of our knowledge, we are the first to explicitly introduce and define the concept of null-variable software. • All the data and code to replicate our study are available at https://github.com/ternava/zero-variability . The remainder of this paper is structured as follows. We first provide the background ( Section 2 ) and motivation ( Section 3 ). Then, we present our empirical study ( Section 4 ) with four research questions, examining the size and variability of over 108 programs. We report on the variability, correlation with system size, evolution, and reduction in Section 5 , and discuss the results in Section 6 . In Section 7 we define null software variability , threats to validity in Section 8 , related work in Section 9 , and conclude in Section 10 . 2. Background Variability-intensive software systems. Software systems that are built nowadays require and exhibit variability, commonly understood as the ability of a software system or artifact to be efficiently extended, changed, customized, or configured for use in a particular context ( 10 ; 43 ) . Software variability is often studied in the context of software reuse ( 33 ) , software factories, software families, software product lines ( e.g. , ( 20 ; 10 ; 6 ) ), and configurable systems ( e.g. , ( 15 ; 46 ) ). The last two abilities, i.e. , customizability and configurability , are typically described in terms of features or variation points with variants incorporated into their respective feature or decision models ( 20 ; 13 ; 48 ) . Similarly, configuration options are used in configurable software systems, turning them into families of closely related software systems ( 15 ) . Moreover, all software artifacts, such as architecture, design, and code, have the potential to vary or to be reused, whether through ad-hoc or systematic reuse. A recent study shows that code artifacts are the most frequently reused ones ( 11 ) . Therefore, in this work, we focus on exploring variability of small-scale configurable