SPLASH/OOPSLA @ 40 Panel

Celebrating the 40th Anniversary of ACM’s SPLASH/OOPSLA Conference Series:

Escaped from the Lab: ACM’s SPLASH Influences on the Future of Software?

Moderator:

• Steven Fraser, Innoxec

Raconteur:

• Dennis Mancl, MSWX

Panelists:

• Jonathan Aldrich, CMU – General Chair SPLASH 2015
• Kent Beck, Thinkies – Program Chair OOPSLA 1989
• Crista Lopes, UCI – Program Chair OOPSLA 2013, General Chair OOPSLA 2011, and Onward! Chair OOPSLA 2007
• Doug Schmidt, William & Mary – Program Chair OOPSLA 2004
• Guy Steele, Tech Squares – Program Chair OOPSLA 2003, Onward! Essays Chair 2015
• Mira Mezini, TU Darmstadt – Program Chair OOPSLA 2023

Escaped from the Lab: ACM’s SPLASH Influences on the Future of Software?

New technologies continue to transform the landscape of programming languages and software engineering. The past 40 years have seen advances in hardware with the evolution of global high speed/capacity networking, mobile devices, IoT, and data centers. Technology innovations have inspired the emergence of software platforms such as web APIs, cloud computing, microservices, and LLM-based AI. Where will we go next? Application programming is no longer limited to the educated, trained, and experienced professional. Users now craft their own spreadsheets, scripting tools, and AI prompt engineering interfaces to define new algorithms, workflows, and automation. In the future, how will SPLASH and similar conferences contribute to catalyzing collaborations, making connections, disseminating research results, and sharing best practices with our dynamic community.

Jonathan Aldrich

Position Statement

Forthcoming.

Bio

Jonathan Aldrich is a Professor of Computer Science at Carnegie Mellon University. He teaches courses in programming languages, software engineering, and program analysis for quality and security. Aldrich directed CMU’s Software Engineering Ph.D. program from 2013-2019.

Aldrich’s research centers on programming languages and type systems that are deeply informed by software engineering considerations and human factors. His research contributions include verifying the correct implementation of an architectural design, modular formal reasoning about code, and API protocol specification and verification. His notable awards include an NSF CAREER award (2006), the Dahl-Nygaard Junior Prize (2007), the DARPA Computer Science Study Group, and an ICSE most influential paper award (2012). He served as general chair (2015), program chair (2017), and steering committee chair (2017-2019) of SPLASH and OOPSLA. Aldrich holds a bachelor’s degree in Computer Science from Caltech and a Ph.D. from the University of Washington. 

Kent Beck 

Position Statement

Like the great trading centers of the ancient world, OOPSLA became the cross-roads where people with overlapping interests came together – academics and practitioners, various schools of programming language design, various scales of teams and forms of risk. OOPSLA/SPLASH has been a supportive place to bring half-baked ideas in hopes of further baking.



Bio

Beck launched his public programming career with a disastrous demonstration of Smalltalk at the first OOPSLA in Portland (1986). Notable achievements include patterns (with Ward Cunningham), JUnit (with Erich Gamma), TDD, and Extreme Programming. Beck is the author of many books on software development and Extreme Programming, including the classic 1999 book “Extreme Programming Explained.” He currently programs daily with the genie and publishes at TidyFirst.substack.com. 

Crista Lopes

Position Statement: The Post-Syntax Era and the Rise of Software “Curation and Verification”

For decades, SPLASH and OOPSLA served as high priests of a “Professional Guild” of software. We defined formalisms, types, and abstractions that have made software manageable for the highly trained few. But the “Lab” has not just been escaped; it has been wrapped into oblivion. When an LLM can generate a Continuation-Passing Style interpreter or a complex web API from a few lines of natural language, the traditional barriers of unintuitive code syntax and skilled code crafting have effectively dissolved.

However, programming is more than the act of writing code; it is the act of precise thinking.

As software creation becomes a ubiquitous end-user activity, the role of conferences like SPLASH must evolve. We are moving from yesterday’s world of Software Construction to tomorrow’s world of Software Curation and Verification. My position for this panel is centered on three shifts:

• From Syntax to Semantics: If anyone can generate code, the value of the “professional” shifts from knowing how to write to knowing what is right. SPLASH must become the vanguard of practices and tools for automated verification and explanation. Our research shouldn’t just be about better languages, but about languages that can “audit” the probabilistic outputs of AI and keep them real.
• The New “Exercises in Style”: Software quality is no longer just about the elegance of expression, but the robustness of intent. We need to define new “Programming Styles” for the AI era – how do we structure conversations with machines to ensure the resulting sociotechnical systems are resilient and secure?
• The Conference as a collaborative sandbox: In a world where research “escapes” to GitHub in hours rather than years, SPLASH’s role is to be the Intellectual Anchor. We shouldn’t just disseminate results; we should be the space where a dynamic community of practitioners and a formal community of theorists come together to define the constraints of this chaotic new reality.

The future of software is messy. Our job is to ensure that this mess will not lead to a Software Dark Age of unmaintainable, hallucinated systems, but rather to a new era of empowered, high-integrity creation.

Bio

Cristina (Crista) Lopes is a Professor of Informatics in the School of Information and Computer Sciences at the University of California, Irvine, and Director of the Institute for Software Research at UC Irvine. Her research focuses on software engineering and programming languages for large-scale data and systems. Early in her career, she was a founding member of the team at Xerox PARC that developed Aspect-Oriented Programming. Along with her research program, she is also a prolific software developer. Her open source contributions include being one of the core developers of OpenSimulator, a virtual world server. She is also a founder and consultant of Encitra, a company specializing in online virtual reality for early-stage sustainable urban redevelopment projects. Her book “Exercises in Programming Style” has gained rave reviews. ACM Computing Surveys chose this volume in their list of “Notable Books” in 2014.
She has a Ph.D. from Northeastern University, and MS and BS degrees from Instituto Superior Tecnico in Portugal. She is the recipient of several awards, including a prestigious CAREER Award and a Pizzigati Prize for Software in the Public Interest.

Mira Mezini

Position Statement

Escaped from the Lab: What OOPSLA/SPLASH Has Given Software – and What It Still Owes. Forty years ago, a small group of visionaries gathered in Portland with a deceptively simple idea: that programming languages and systems could be designed with greater expressiveness, humanity, and elegance. OOPSLA became a crucible for some of the most consequential innovations in the history of computing.

The energy of those early years was electric. When I arrived as a young Ph.D. student in 1994 — working on making the meta-level architecture of Smalltalk more expressive — I was overwhelmed by the community’s vitality. Over 3000 participants filled the halls; the industry exhibition floor brought together researchers and industry practitioners in genuine, productive dialogue. Industry was not merely consuming research, it was shaping it. We should recapture that symbiosis.

In the early days, OOPSLA was focused on Smalltalk and C++, but it also set in motion a cascade of innovation. Live programming environments, Model-View-Controller, reflective architecture led to environments like Self that incorporated virtual machines. Today we see that power in JavaScript engines and the JVM. We also developed metaobject protocols, open implementations, and meta-level architectures. I found myself recommending Kiczales’s OOPSLA 1994 keynote to my students as we were shaping the vision of application-driven systems.

The cascade of innovations continued well beyond object-orientation. OOPSLA became the venue that advanced many approaches to software production: design patterns, refactoring, aspect-oriented programming, test-driven development, agile methodologies, wikis, and model-driven engineering. New language ideas were refined at OOPSLA: type systems, module systems, and compiler techniques for dynamic languages. An important idea today is ownership types, introduced in a 1998 paper by Clarke, Potter, and Noble (“Ownership Types for Flexible Alias Protection”). This paper asks how a type system can enforce that objects do not alias across encapsulation boundaries. Decades later, Rust’s ownership and borrow checker system draws directly on this intellectual lineage. Few stories better illustrate OOPSLA’s mode of impact: a formal type-theoretic paper from 1998 that was born in our community quietly became the foundation of one of the most celebrated programming languages of the 2020s.

All of these ideas escaped the lab — which is, of course, the apt title of this panel — and colonized the software industry in ways their authors could not always foresee.

“Vibe coding” is connected to a 2003 Onward! paper “Beyond AOP: Toward Naturalistic Programming” by Cristina Lopes, Paul Dourish, and colleagues. I have sponsored follow-up work in my lab on “Towards Naturalistic Programming.” What if programs could be written in terms closer to natural human thought rather than formal notation? Today’s “coding through AI agents using natural language prompts” is making a kind of naturalistic programming real — but in the most ad hoc, uncontrolled, and semantically impoverished way imaginable. The vision our community articulated was not wrong; the execution happening in the wild is simply happening without us. And that is a problem.

The most transformative shift now reshaping how software is written — the rise of large language models for code generation — has been designed almost entirely by machine learning researchers, not by programming language designers, type theorists, or compiler engineers. In 2023, Dimitrios Vytiniotis’s keynote talk explored the intersection where PL research and ML systems meet. His talk “Scaling Up Machine Learning Without Tears: And What Do Programming Languages Have to Do with It?” has started an important conversation. But one keynote is not enough. The broader PL and software engineering community has largely remained a bystander while stochastic systems generate billions of lines of production code that cannot be formally verified, whose correctness no one can guarantee, and whose trust deficit grows by the year.

The naturalistic programming vision that Onward! dared to articulate twenty years ago needs to be rescued from the current ad hoc reality and given the rigor it deserves: grounded semantics, compositional correctness, and the kind of principled language design that has always been OOPSLA’s hallmark. Future SPLASH conferences could be defined by this challenge. I hope we are bold enough to take it on.

Bio

Mezini is a Professor of Computer Science at TU Darmstadt (TUDa), leading the Software Technology Lab. After completing her Ph.D. at the University of Siegen in Germany, Mezini was an assistant professor at Northeastern University (USA) before joining TUDa in 2000. She serves on the board of the German National Research Center for Applied Cybersecurity ATHENE and co-directs hessian.AI, the Hessian Center for Artificial Intelligence. Mezini held various leadership positions at TUDa, including Dean of Computer Science (2013-2014) and Vice President roles (2014-2019). She has held visiting professorships at Lancaster University (UK) and Università della Svizzera Italiana (Switzerland).

Mezini’s research focuses on three areas: programming systems for reliable distributed software and AI, automated software analysis, and foundational code models. Her awards include two IBM Eclipse Innovation Awards (2005 and 2006), a Google Research Award (2017), and the second prize in the Horst Görtz Foundation’s IT Security Award (2014). A member of the German Academy of Engineering Sciences and the Academia Europaea, Mezini was recently named an ACM Fellow and elected to the National Academy of Sciences Leopoldina, further cementing her status as a leader in the field of computer science. Mezini was awarded the Dahl-Nygaard Prize 2025. The prize is one of the most prestigious in the field of programming and software engineering. She is also co-spokesperson of the Excellence Cluster “Reasonable Artificial Intelligence”. 

Douglas C. Schmidt

Position Statement

Over the past four decades, OOPSLA – and later SPLASH – has played a key role in shaping how the software community thinks about abstraction, composition, and evolution. At its best, OOPSLA was never merely a venue for programming language features or tools; it was a forum where ideas about software as a human activity were debated seriously. Concepts such as modularity, reuse, frameworks, software patterns, and middleware were not treated as academic curiosities, but as responses to real scalability pressures in software-reliant systems. This pragmatic–theoretical balance distinguished OOPSLA from both purely academic programming language venues and purely industrial conferences.

A defining contribution of OOPSLA was its insistence that software architecture and developer cognition matter as much as formal correctness. The goal of human understanding of software has led to work on patterns, reflective systems, component frameworks, and object-oriented middleware. This valuable research bridged programming languages, systems, and software engineering. OOPSLA’s intellectual “middle ground” enabled ideas to move quickly from papers to operational platforms, influencing mainstream practice in enterprise systems, distributed computing, and later cloud-native architectures. In many ways, OOPSLA normalized the idea that good software research must engage with scale, deployment, and evolution, rather than elegance or formalism alone.

From a long-term perspective, the enduring influence of OOPSLA/SPLASH lies less in any specific language or paradigm and more in advancing three norms of software production. First, software abstractions must earn their keep in real systems. Next, developer productivity and system evolution are first-class research concerns. Finally, dialogue between academia and practice is not optional. Although generative AI and intent-driven development are reshaping how today’s software is built, the original OOPSLA ethos – questioning how abstractions amplify or constrain human intent at scale – feels newly relevant. The challenge for SPLASH going forward is to reclaim that integrative role. SPLASH must once again serve as the place where the software community asks its hardest, most consequential questions. We should not chase trends.

Bio

Schmidt is the inaugural Dean of the School of Computing, Data Sciences, and Physics at William & Mary and a longtime leader at the intersection of software research, practice, and policy. He previously held senior academic and research leadership roles at Vanderbilt University and the Software Engineering Institute at Carnegie Mellon University. Over four decades, his research has shaped object-oriented frameworks, software patterns, middleware, and model-driven engineering for large-scale, mission-critical systems, with recent work focusing on prompt engineering patterns and generative augmented intelligence. He holds degrees from William & Mary and the University of California, Irvine and has been an active contributor to the OOPSLA/SPLASH community throughout his career. 

Guy L. Steele Jr.


Position Statement

The first major conference sponsored by SIGPLAN that turned into an annual series was POPL, starting in 1973, and it has remained much the same for more than half a century now, focusing on the principles of programming language design. The second major SIGPLAN conference series started in 1979 as a specialty symposium on Compiler Construction, held every two to three years, but in 1988 it became PLDI, centered more broadly on all the practical aspects of programming language design and implementation. So if you were writing programming-language papers in the late 1980s and early 1990s and wanted to publish in a SIGPLAN conference, the decision process was pretty simple: if your paper was theoretical and of broad interest, try POPL; if it was practical and of broad interest, try PLDI; otherwise look for a specialized workshop or conference, such as LFP (Lisp and Functional Programming), ASPLOS, or PPoPP.

One of those newly created specialized conferences was OOPSLA (1986). Object-oriented programming, epitomized at that time by Smalltalk, was beginning to attract broader attention, and there were many experiments in adopting object-oriented programming ideas into existing or new programming languages. The definition of C++ had just been published the year before. New object-oriented languages and ideas published at the first OOPSLA conference touched on over a dozen different programming languages and environments.

There was clearly so much cooking in this new arena of object-oriented programming that OOPSLA, unlike PLDI and ICFP (which grew out of the biennial Conference on Lisp and Functional Programming), immediately became an annual conference.

And then a funny thing happened: in the 1990s and into the 2000s, the center of gravity of programming language design shifted away from “conventional” procedural languages (e.g., Fortran, Algol, Pascal, C, and Ada), towards object-oriented programming languages such as C++, Ruby, Java, JavaScript, Python, and all their successors. Today, object-oriented programming is not a specialty topic; it is entirely mainstream, and this has been reflected by the use of “SPLASH” as the umbrella conference title since 2010. One could argue that SPLASH has become SIGPLAN’s broadest conference, encompassing more topics than POPL, ICFP, or PLDI. All these conferences are important, but SPLASH now has a much more central role than I might have anticipated at the turn of the millennium.

Four topics that may be getting ever-increasing programming-language attention in the near future are (1) quantum computing, (2) homomorphic encryption, (3) programmable massively-multiplayer online games, and (4) language language models and related AI applications. The last of these may change the very nature of the programming process itself in ways that have already begun to unfold and are hard to predict. Nevertheless, I suggest that SPLASH is likely to continue to be an appropriate venue for publication in all of these areas as they develop – yes, POPL, PLDI, ICFP, and ASPLOS will also have roles to play as the theory and technologies mature, but I suspect SPLASH will continue to be a melting pot that encourages adventurous preliminary exploration as well as reporting on successful deployment.

Bio

Steele retired in January 2025 after a career of more than 50 years in the design, documentation, implementation, maintenance, and standardization of computer programming languages. He joined ACM in 1971; the same year he implemented a Lisp interpreter for the IBM 1130.

He taught at Carnegie-Mellon University, then worked for Tartan Laboratories, Thinking Machines Corporation, Sun Microsystems Laboratories, and Oracle Labs. He has been given the Grace Murray Hopper Award (1988), the ACM SIGPLAN Programming Languages Achievement Award (1996), and the Harry H. Goode Memorial Award (2007). He is an ACM Fellow, IEEE Fellow, Fellow of the American Academy of Arts and Sciences, and a Member of the National Academy of Engineering of the United States of America.

He has served on the accredited standards committees X3J11 (C language) and X3J3 (Fortran), and additionally served as chairman of X3J13 (Common Lisp). He served as the first project editor for the ECMAScript standard for JavaScript. He and Gerald Jay Sussman created the Scheme programming language, and he was a member of the IEEE committee that produced IEEE Standard 1178-1990 for Scheme.

In 2021, he served as co-chair of the Fourth ACM SIGPLAN Conference on the History of Programming Languages.

Steven D. Fraser

Position Statement

From my first OOPSLA in 1989 (and 25+ since then), I’ve seen an evolution in conference purpose and community. As a conference, OOPSLA initially brought together academic researchers and industry practitioners to discuss, debate, and learn. That is still the case today, however it seems with more academics than industry practitioners than in the past. Various conference elements have changed since the early days. No longer is there a tools “trade show” or a tutorial offering. It also appears to be increasingly challenging to obtain corporate funding. In-person conferences continue to face increasing costs (travel and venue) and administrative (e.g., government visa and conflict) challenges. Perhaps new virtual interaction paradigms and communication platforms will emerge to make hybrid conferences more attractive (and adopted) – catalyzing and inspiring new global collaborations and innovations.

Bio

Fraser is a strategic advisor and frequent conference speaker on tech transfer, company-academic R&D partnerships, and agile software development practices. Based in Silicon Valley, his work is focused on partnerships to accelerate product R&D through collaboration, networking, and “open innovation” ecosystems. As a conference panel “Impresario,” since 1994 he has organized and moderated over 75 panels and workshops for ACM’s SPLASH/OOPSLA, IEEE’s ICSE, and the Agile XP Conference Series. Prior to Innoxec, Fraser led HP’s Global University Programs and served as the Director of the Cisco Research Center orchestrating university research investments, leading PhD recruiting, and catalyzing company collaborations through company-wide forums. Fraser was a Senior Manager (Global External Research) at BNR/Nortel and a Visiting Scientist at Carnegie Melon University’s Software Engineering Institute (SEI) prototyping and piloting team-based domain engineering techniques to leverage commonality in software product families. Fraser received a Ph.D. in Electrical (Software) Engineering from McGill University (Montréal), a M.Sc. in Engineering Physics (Queen’s at Kingston), and a B.Sc. in Physics and Computer Science (McGill). Fraser is a Senior Member of both the ACM and the IEEE.

Dennis Mancl

Position Statement

In its forty years, OOPSLA and SPLASH started from an early showcase for a useful but underdeveloped software design approach, then developed into an important forum for exploring new software engineering ideas, and finally became a place to critique and reflect on how we learn and teach new software technologies. Software isn’t dead yet – we refuse to be replaced by robots and smart-aleck machine learning technology. We have learned that both research and commerce will be better served by relying on the knowledge, experience, and judgment of living people – the best path for software continue to serve humanity.

Bio

Dennis Mancl is a retired software developer and coach, with decades of Bell Labs experience in the area of software practices in telecom. Mancl is especially interested in “soft” issues in software engineering: teamwork, collaboration, design brainstorming, refactoring, diverse teams, critical thinking skills, and continuing education. Most of his current work is about how to be a good lifelong learner. He is still an occasional presenter at international ACM and Agile conferences, where he helps organize and report on conference panel discussions. Mancl has been a volunteer with the Princeton Chapter since 1983. Mancl shares some of his recent presentations and other research at https://manclswx.com.