On the Shoulders of Giants

Collaboration, Reuse, and Recognition in R&D.

On my resume sits a quote recorded by John of Salisbury in the 12th century:

Bernard of Chartres used to compare us to dwarfs perched on the shoulders of giants. He pointed out that we see more and farther than our predecessors, not because we have keener vision or greater height, but because we are lifted up and borne aloft on their gigantic stature.

β€” Bernard of Chartres (recorded by John of Salisbury in Metalogicon, 1159)

This quote is not just a philosophical tagline; it is a description of the mechanics of progress in both scientific research and software engineering. Whether navigating academic state-of-the-art analysis or industrial software shipping, recognizing that we do not build in a vacuum is a superpower.


1. The Humility of the State of the Art

In academic research, this concept is highly codified. Before a researcher can develop anything new, they must spend months (sometimes years) conducting literature reviews to establish the state of the art.

Peer reviews and comparative benchmarks are not bureaucratic hurdles; they are the literal manifestation of standing on the shoulders of giants. When we compare our work against existing methods, it is not about proving ourselves superior in a vacuum. It is about intellectually positioning our work as a continuation of a larger dialogue. Acknowledging that our β€œnovel contribution” is often just a small, incremental layer on top of a massive mountain of historical research is what keeps research honest, rigorous, and collaborative.


2. The Cost of Reinventing the Wheel

In software engineering, this concept is manifested in open-source software, standard libraries, and design patterns. Yet, many software engineers fall victim to the β€œNot Invented Here” (NIH) syndrome: the temptation to write custom mathematical utilities, geometry parsers, or neural network wrappers from scratch because they believe they can build a β€œcleaner” or β€œmore optimized” version.

The hidden cost of reinventing the wheel is enormous. It is not just the time spent writing the initial code; it is the lack of battle-testing. A custom-rolled utility has zero exposure to real-world edge cases. An established open-source library, however, has been tested, debugged, and optimized by thousands of developers over several years.

By refusing to stand on these existing engineering shoulders, we waste time debugging basic plumbing instead of focusing on the unique, high-value problem we are actually trying to solve.


3. Occam’s Razor: Simplicity Through Familiarity

This philosophy aligns directly with Occam’s razor, the principle that the simplest solution is usually the best. In engineering, the simplest solution is rarely the one we write from scratch; it is the one that introduces the least amount of custom code and surface area for bugs.

Relying on existing work is the ultimate application of Occam’s razor. By choosing standard, well-documented libraries and algorithms, we keep the codebase clean and highly readable. If a new engineer joins the project, they don’t have to spend weeks learning a custom, home-grown utility; they can rely on their pre-existing knowledge of standard tools. Simplicity isn’t just about how short the code is; it’s about how familiar, maintainable, and predictable it is.


4. Collaboration, Recognition, and Team Intersections

There is a common misconception that leveraging the work of others, whether open-source authors or colleagues in a neighboring department, somehow diminishes our own contribution.

In reality, the opposite is true. True innovation is rarely a solitary act. In industrial R&D, success depends on the intersection of diverse skill sets. To build something truly innovative, research scientists must collaborate closely with core software developers, product teams, and infrastructure engineers.

Building upon a colleague’s work is not a compromise of our intellectual independence; it is a prerequisite for collective success. When we recognize and respect the foundations laid by other teams, we don’t dilute our impact; we amplify it. By spending our energy bridging our respective expertise rather than recreating each other’s work, we construct solutions that would be impossible to build in isolation.


5. Conclusion

Humility and pragmatism are the hallmarks of a mature researcher and engineer. The next time you begin a project, do not seek to prove how much you can write from scratch. Instead, focus on how much you don’t have to write because you successfully stood on the shoulders of the giants and the colleagues around you.

Arnaud Bletterer

Arnaud Bletterer

Research Scientist & Team Leader

Specializing in Computer Graphics, Image/Geometry Processing, and Machine Learning, with a passion for bridging the gap between mathematical exploration and robust industrial software. If you have thoughts on these topics, are facing similar R&D challenges, or simply want to exchange ideas, feel free to reach out.