Thanks to Rich

Richard DeMillo deserves, in my opinion, an award for his decades of research. A difficulty I believe is that he has worked on multiple areas and made important contributions to each of these areas. Let’s take a look at the top two.

For the record he held prior to joining Georgia Tech: Chief Technology Officer for Hewlett-Packard, Vice President of Computing Research for Bell Communications Research, Director of the Computer Research Division for the National Science Foundation, and Director of the Software Test and Evaluation Project for the Office of the Secretary of Defense. He has also held faculty positions at the University of Wisconsin, Purdue University and the University of Padua, Italy.

What Deserves Recognition?

A basic question is what research should be recognized? Solving a longstanding open problem—P vs NP—would immediately need to be awarded. But another critical category is an idea that leads to an interesting interplay between practical and theoretical work. Something that both advances our understanding of part of computing but also is used to solve real problems. Especially problems that are important and need to be solved to advance society.

In my opinion, for Rich, program testing is one such area and also cryptography is another. Let’s turn and look at these now.

Program Testing

The method called program mutation is what deserves recognition. It has been around since the 1970’s. The reasons it should be given recognition are simple:

• It is identified in more than 400 papers published in the time
  period 2008—2017;

• It is featured in its own conferences;

• It is used in 87 different mutation tools for a variety of
  programming languages and artifacts including Java;

• It is used in practice.

I conceived the idea in a student paper in 1971, but Rich had a huge role in actuating it. This came in papers with me, him, and Fred Sayward in the late 1970s, which helped Tim Budd create the first implementation for this 1980 PhD thesis. Besides what Ken and I recently wrote about the history, we note the following from a survey dated last July:

Since its introduction back in the 70s (DeMilo et al. 1978, 1979),
research on mutation testing has thrived until becoming a
well-established testing technique. In a recent survey by Papadakis et al. survey, the authors identified more than 400 papers published in the time period 2008—2017 and 87 different mutation tools for a variety of programming languages and artifacts including Java, C, C++, C#, JavaScript, HTML/CSS, Ruby, and UML models, among many others.

For almost two decades there has been a whole conference on mutation analysis. A public website and blog by Dietrich Geisler of Cornell includes a graph that illustrates the general growth of mutation testing as a topic:

Here are some more links:

• A paper in 2000 by Jeff Offutt and Roland Untch;

• A survey in 2022 by Ana Sanchez, Pedro Delgado-Perez, Inmaculada
  Medina-Bulo & Sergio Segura;

• A recent paper on over a hundred mutation projects.

In the 1980’s DeMillo advised Offutt who was his PhD student.

Security

DeMillo is a co-inventor of Differential Fault Cryptanalysis and holds the patent on applying DFA to break public key cryptosystems. On the Importance of Eliminating Errors in Cryptographic Computations.

This is related to mutation testing, since it is a kind of change the code method. Eli Biham and Ali Shamir say:

We would like to gratefully acknowledge the pioneering contribution of Boneh, DeMillo, and Lipton. whose ideas were the starting point of our new attack.

Here are some other papers.

Open Problems

Perhaps we can see our way to help recognize Rich? For the record: Rich is a professor at Georgia Tech’s School of Cybersecurity and Privacy. He holds the Charlotte Brody and Roger Warren Chair in Computing at Georgia Tech. He is is also Managing Director of Gtatrium, LLC, a subsidiary of Georgia Advanced Technology Ventures. He was formerly the John Imlay Dean of Computing and Director of the Georgia Tech Information Security
Center.

[fixed jaggies]