Skip to main content

Congratulations to the award, César!

"Receiving this award has been a tremendous achievement in my career"

Photo: Fredrik Persson
Published Oct 26, 2021

César Soto Valero has won a paper award in a top software engineering conference for his research on mitigating software bloat. Software bloat is code present in software applications, that is not necessary to execute the applications correctly. Instead, it is a pervasive problem that wastes computational resources, creates maintenance issues, and compromises security.

We have spoken to César – a doctoral student who is part of the EECS research centre CASTOR  – about the paper, his research and what impact it might have.

​What does this win mean for you and your research?

"Receiving this award has been a tremendous achievement in my career. I am honoured to share this result with two world-class researchers: my co-supervisor Thomas Durieux , and supervisor Benoit Baudry.  I see this award as a recognition from the research community for the hard work we have carried out this year. It confirms the quality and impact of our research and helps me stay motivated and focused on my current goals."

Tell us a bit about your research and the impact it can have on society.

"My research is directed towards mitigating software bloat through static and dynamic program analysis techniques. Software bloat refers to code present in software applications, but that is not necessary to execute the applications correctly. The presence of software bloat is a pervasive problem today because it needlessly wastes computational resources, creates maintenance issues, and compromises security."

"Our paper is the first work that studies software bloat over time to understand this phenomenon. To this end, we took a deep look at one particular type of software bloat, which we coined as "bloated dependencies." We empirically demonstrate that when one specific software dependency becomes bloated, then it is very likely that it will remain bloated in the future. This finding has important practical connotations because developers are often sceptical of adopting debloating tools since they wonder if they might need a piece of bloated code in coming releases."

"A second significant contribution of our paper is determining if developers actively maintain bloated dependencies in popular software projects. Interestingly, developers often update bloated dependencies, and popular bots perform this activity regularly, which is a considerable waste of maintenance effort that can be minimised by removing these dependencies. Armed with this novel knowledge, software engineers can be more confident in removing a bloated dependency. Our results encourage developers to pay more attention and take action to mitigate this phenomenon."

"This work is in line with the objectives of the CASTOR research centre. In particular, when it comes to researching the highest scientific standards while providing open-source tools and fostering collaborations between academia and industry. We are actively working with our founding partners to implement our debloating tools in their software systems."

What is the most exciting research happening in your field?

"I believe we are at a very early stage regarding the way we produce and consume software technology. The current software development tools that we use today are very primitive. Soon, the adoption of advanced AI-powered tools for software development will become ubiquitous. Human developers and bots will cooperate to create complex software systems. With this technology, developers will perform advanced code analysis and refactoring tasks that today seem practically impossible. A lot of fascinating research is happening in this direction."

"On the other hand, the world is moving more and more towards adopting decentralised applications and services. However, the current implementations of blockchain are far from perfect. Exciting research is taking place beyond blockchain in an attempt to overcome its limitations. Overall, a large body of software engineering research is devoted to developing better mechanisms for hardening the software supply chain. Urgent needs from the industry directly drive this. Therefore, significant research effort is underway to make existing applications faster, more secure, and easier to maintain."


Here are his contact details.