In this episode of the a16z Crypto Show, host Tim Roughgarden (Head of Research at a16z Crypto) and research partner Itay Abraham sit down with Turing Award-winning computer scientist and MIT professor Dr. Barbara Liskov. The conversation explores the fundamental computer science concepts that underpin modern blockchain protocols, tracing Dr. Liskov's pioneering career from programming language design to Distributed Systems.

The speakers discuss the chronological evolution of replication protocols, beginning with Viewstamped Replication in the 1980s and its transition to Practical Byzantine Fault Tolerance (PBFT) in the late 1990s. Dr. Liskov details how her early work on modularity and data abstraction in languages like Clue and Argus informed her approach to solving consensus and fault tolerance in distributed networks. Finally, the discussion addresses the future of systems research and how the rise of artificial intelligence is reshaping the role of software engineers, emphasizing the enduring importance of first-principles thinking, verification tools, and high-level system design.

The Evolution from Programming Languages to Distributed Systems

Dr. Barbara Liskov transitioned to Distributed Systems in the early 1980s after reading a paper by Bob Kahn about the unrealized dream of distributed computing.

"I had thought a little bit about maybe starting a company to market the language, but I realized that I was much more interested in research." - Barbara

The Argus programming language extended Clue by introducing "guardians," which were concurrent objects residing at single network nodes that communicated via atomic transactions.

"Modularity is everything in building large programs... you can do a proof of the correctness of the code that implements the module without having to look at any other module's code." - Barbara

Viewstamped Replication and Solving Benign Failures

Developed in the mid-1980s with student Brian Oakey, Viewstamped Replication aimed to build a reliable, replicated file system that did not rely on client-side locking.

The protocol solved the "window of vulnerability" in two-phase commit by allowing backups to transition to a new "view" and elect a new primary if the original failed.

"We came up with a protocol that if the primary seemed to not be doing its job, the backups then carried out another protocol in which a different replica became the primary." - Barbara

The protocol assumed benign failures where machines were either running or silent, and corrupted messages could be easily detected, which was typical for the early ARPANET era.

Practical Byzantine Fault Tolerance and Malicious Attacks

In the late 1990s, inspired by a DARPA request for proposals on handling internet attacks, Dr. Liskov and her student Miguel Castro developed Practical Byzantine Fault Tolerance (PBFT).

"Why don't we see whether we can figure out a way to do replication that handles these malicious attacks?" - Miguel

While benign failure models require 2F+1 replicas to tolerate F failures, Byzantine models require 3F+1 replicas to handle nodes that actively lie or send conflicting messages.

PBFT introduced a three-phase protocol using cryptographic certificates of 2F+1 signed messages to prove consensus on state transitions without trusting any single replica.

"The way you handle the problem is you never trust an individual replica. You only trust the group, and the group has to consist of a sufficient number of replicas to be able to prove that this is what really happened." - Barbara

State Machine Replication as the Blockchain Foundation

State machine replication separates the consensus layer (ordering operations in a ledger) from the execution layer (interpreting what those operations mean).

"We just wrote into the ledger what the operation was that was being requested, but we were not in the least bit interested in what it meant to execute that operation." - Barbara

Modern Turing-complete blockchains like Ethereum and Solana are literal implementations of fully general state machine replication protocols.

Although Bitcoin initially used a different consensus mechanism, the blockchain industry eventually recognized PBFT-style protocols as the optimal foundation for many decentralized systems.

The Future of Systems Research and Coding in the AI Era

Dr. Liskov expresses concern about the limits of AI-generated code, noting that verification tools will be critical to ensure correctness.

"I can easily see how AI can write a little program for you if you give it a specification. I can easily see how that program might be wrong." - Barbara

Future software engineers must shift their focus from low-level syntax to high-level system design, modularity, specifications, and verification.

"Students really need to understand how to write those programs themselves, or otherwise, how will they know that AI did them right or wrong?" - Barbara

Resources Mentioned

The Rise of AI How Artificial Intelligence is Transforming Work, Privacy, and the Future of Society

nd fault tolerance in distributed networks. Finally, the discussion addresses the future of systems research and how the rise of artificial intelligence is reshaping the role of software engineers, em

The Augmented Dentist Clinical Precision and Practice Growth in the AI Era (The Next-Gen Practice Blueprint Book 4)

FT-style protocols as the optimal foundation for many decentralized systems. The Future of Systems Research and Coding in the AI Era Dr. Liskov expresses concern about the limits of AI-generated code

in that area

d recognized that this was a serious problem, the problem of malicious attacks, and was looking for research in that area. I had a student, Miguel Castro. He came to me and he said, "Why don't we see

is concerned— As part of our summer feed drop

Computer science is in a very strange place right now with the advent of AI. I think that as far as research is concerned— As part of our summer feed drop, series, we're sharing another episode from t

Crypto for Absolute Beginners Workbook Practical Exercises and Checklists to Help You Learn Cryptocurrency With Confidence

ome to First Principles, a series by the team at a16z Crypto. I'm Tim Roughgarden. I'm the head of research at a16z Crypto. And today we're exploring a core challenge in distributed computing. How ca

partner Itay Abraham joins me as we talk with Dr

ication from the 1980s to practical Byzantine fault tolerance, or PBFT, a decade later. a16z crypto research partner Itay Abraham joins me as we talk with Dr. Liskov about her path from programming la

as I had anything interesting to do at that point

ore interested in research. And I also thought that I had sort of done as much programming language research as I had anything interesting to do at that point. And I started looking around and I read

Elan Publishing Company Field Notebook/Pocket Journal - 3.5"x5.5" - Black - Lined Memo Book - Pack of 5

ubt about it. I'm curious, did it feel like a pivot at the time, like you were really changing your research field? And maybe a related question would be, you know, how much did sort of the mindset an

Business Essentials for Healthcare Professionals How to Operate a Sustainable, Profitable, and Salable Practice or Successfully Work for Someone Else.

cas and where lots of people could be using the same files. In those days, if you wanted to write a paper or do other work with somebody else, you had to use file transfer and, you know, you couldn't

Us Against You A Novel (Beartown Series)

maybe a little bit of a digression, but I'm just curious, which is a little bit about what systems research was like at that time as a community. So my career started kind of around 2000 or so. And f

are very closely connected because when you teach well

uh, MIT is a wonderful place and we also took teaching very seriously. And I do think teaching And research are very closely connected because when you teach well, you teach from first principles. A

BOOK I THINKING FROM FIRST PRINCIPLES Understanding Reality Before Models (Thinking Systems for an Uncertain World A First, Second, and Third Principles Journey into Modern Intelligence 1)

st principles. And when you do good research, you have to truly understand. So you're really doing research from first principles. And it's also very important in research to understand what you don'

Too Much Information Understanding What You Don't Want to Know

understand. So you're really doing research from first principles. And it's also very important in research to understand what you don't understand. I always tell my students that there's where you g

is the special case where K equals zero

e number of people who understand something from a smaller number K to a bigger number N. Yeah. And research is the special case where K equals zero. Or for the first time people are understanding it

Wild Side

that's kind of the twin pillars of the professor's role, I think, really. Yeah. So returning to the research side, so View Stamp Replication came out in the late '80s. And then there was sort of a per

was published and that paper used replication and talked about using Paxos

understanding. Yeah. The story I heard was, well, in the late, in the '90s, the Google File System paper was published and that paper used replication and talked about using Paxos. And I was kind of

was getting funded all those years

I think that DARPA was a very important— between DARPA and NSF, I mean, they were the reasons that research was getting funded all those years. They are one of the very important backbones of why we

Kingdom Funding How To Secure Grants and Government Funding

re one of the very important backbones of why we have the internet that we have today. Without that research funding, it would've been much more difficult to have done all that work. So that's how we

and all of the reactions around it was perhaps there was this aha moment

heoretical work to practical systems, right? I mean, my sense just, you know, from reading the PBFT paper and all of the reactions around it was perhaps there was this aha moment. It's like, oh, wait

Before I Say Goodbye A Novel

, oh, this, this technology can actually be used. There's a difference between having a theoretical paper saying it's possible and seeing, you know, good benchmarks. Oh, I see what you're saying. Say

I’m the Monster Girl You Once Saved A Middle-Aged Teacher in Another World Is Concerned About His Sudden Popularity, Vol. 1

nt of AI and not really understanding, uh, where we're all going with this. I think that as far as research is concerned, there's huge amounts of research to be done, and I can't help but think about

What Is to Be Done?

e we're all going with this. I think that as far as research is concerned, there's huge amounts of research to be done, and I can't help but think about all the systems research that's lying under al

The Alignment Problem Machine Learning and Human Values

rned, there's huge amounts of research to be done, and I can't help but think about all the systems research that's lying under all the AI stuff and all the problems that are coming up that make for v

K Is in Trouble (A Graphic Novel)

l the problems that are coming up that make for very interesting systems research. So I don't think research is in trouble. I am a little worried about computer science as a, as a field for young, you

going on whether this is a field that you can get a good job in and do the kind of coding work that people used to do

hat AI did them right or wrong? And that's correct. I think, you know, there's going to be a lot of research going on whether this is a field that you can get a good job in and do the kind of coding w

Area 51 Earth Abides

o the extent that we can figure out ways to modify or contain the bad behavior, that's another good research area. So what am I saying? I'm saying there's great research to be done and a lot of it's i

to be done and a lot of it's in the systems area

the bad behavior, that's another good research area. So what am I saying? I'm saying there's great research to be done and a lot of it's in the systems area. I couldn't agree more. I think that's a,