text stringlengths 0 2.46k |
|---|
**Jerod Santo:** Perfect. That's the Easy button. Of course, as Adam said, we link up everything in the show notes for y'all, so you can just click through and get where you want to go. Andreas, thanks for coming on. We appreciate our multiple listeners who asked us to have you on the show. You might not have crossed o... |
**Adam Stacoviak:** Gosh, Jerod... |
**Jerod Santo:** \[laughs\] How cool would that be? That'd be the most amazing story. Okay, I'll go 5%. I'll take a 5%. But even if it's just our friends and family, I'd be happy with that as well, but... I love the project, I love hearing about it. It's pretty cool. |
**Andreas Kling:** Yeah, thanks for having me. |
**Adam Stacoviak:** Thanks, Andreas. |
• LaTeX and its history, including its pronunciation controversy and its relation to Don Knuth's TeX typesetting program |
• Leslie Lamport's contributions to LaTeX, including building a macro package on top of TeX |
• The separation of content and presentation in LaTeX, and its significance in document preparation |
• Leslie Lamport's tendency to build on the work of others and sometimes receive credit for new ideas |
• The importance of mathematical thinking in computer science, and how Leslie Lamport approaches problems differently than most computer scientists |
• Examples of how Leslie Lamport's mathematical thinking leads to different solutions and perspectives on problems in computer science |
• Separation of concerns between program ideas and implementation |
• Distinction between coding and programming |
• Importance of formal specification in software development |
• Writing a clear description of what the program is supposed to do |
• Use of pseudocode or informal English to explain complex ideas |
• Benefits of writing code from a formal specification, such as ease of modification and maintenance |
• Contrast with current software development methodologies, such as agile and "move fast and break things" |
• The importance of upfront design and formal specification in software development |
• The dangers of patching code without a solid foundation |
• The value of thinking in algorithms versus writing code |
• The need for a different mindset in programming, thinking outside the box of programming languages |
• The role of math in software development, and the importance of formal methods |
• The challenges of teaching programmers to think mathematically and use formal methods |
• The benefits of using tools like TLA+ to prove the correctness of software before coding it |
• Leslie Lamport's experience with education and learning TLA+ |
• The limitations of traditional teaching methods and the importance of practical application |
• Resources for learning TLA+, including open-source specs and Leslie's own video course |
• Leslie's career and plans for retirement |
• Changes in the industry since Leslie started, including the complexity of software and the impact of evolution |
• The importance of thinking outside the box of code and Leslie's work on distributed systems, including the bakery algorithm and the Paxos paper |
• The threshold for when a system becomes a distributed system and Leslie's definition of distribution |
• Traditional programs vs. concurrent systems |
• Comparing concurrent systems to movies with discrete events |
• State of a system as a movie, with variables and other aspects of state |
• "Stuttering insensitivity" property of TLA+ specfication language |
• Difficulty of describing concurrent systems without stuttering insensitivity |
• Simplifying concurrent system descriptions with stuttering insensitivity |
• Leslie Lamport's bakery algorithm and its relationship to stuttering insensitivity |
• The bakery algorithm is a concurrent algorithm that solves the mutual exclusion problem |
• The algorithm was introduced by Leslie Lamport in response to a complicated mutual exclusion algorithm he found in a paper |
• The algorithm allows multiple processes to access shared variables without the need for mutual exclusion on reading and writing |
• The algorithm's remarkable property is that it does not require mutual exclusion on reading/writing variables, which is a common assumption in concurrent algorithms |
• Leslie Lamport discovered this property when writing the correctness proof for the algorithm |
• The algorithm is based on the idea of people choosing their own numbers and picking a bigger number, similar to how a bakery would assign numbers to customers |
• The algorithm has been formalized and is now written in TLA (Temporal Logic of Actions) |
• Leslie Lamport and Jerod Santo discuss the origins of Lamport's bakery algorithm and how it was discovered and published. |
• The conversation shifts to Lamport's work on concurrent algorithms and how studying the bakery algorithm led to the development of new algorithms and methodologies. |
• Lamport discusses his later work on distributed systems and his contributions to Azure and AWS. |
• The conversation concludes with Lamport discussing his current work on a book about TLA+ and his role at Microsoft, which involves managing a colleague and devoting time to the book. |
**Jerod Santo:** Alright, I'm here with Leslie Lamport, a distinguished scientist at Microsoft Research, creator of LaTeX; not the |
rubbery substance, but the documentation system that you've probably heard of or used. It's used in Academia, it's used in scientific circles, math circles... And a Turing Award winner. Leslie, thanks so much for coming on the show. |
**Leslie Lamport:** Thank you. You might want to try that again and pronounce it either LayTech or LaTec. |
**Jerod Santo:** Ah, I was reading about this controversy, the pronunciation controversy. So as a layman, I see L-A-T-E-X and I do think of that rubbery substance. But the initial author -- it was Tech, right? It was supposed to be T-E-C-H was the sound... And your take on that is LaTech. That's how I should say it? |
**Leslie Lamport:** Or LayTech. Either one. |
**Jerod Santo:** Alright. Now, where do we get Tech out of Tex? That's the curious bit. |
**Leslie Lamport:** Oh, that's because LaTeX is built on top of Don Knuth's tech typesetting program. |
**Jerod Santo:** Okay. And Knuth called it Tech, even though it was Tex, is that right? |
**Leslie Lamport:** Yeah. He says it's the Greek pronunciation of the original word. Not speaking Greek, I wouldn't know. |
**Jerod Santo:** Ah... Well, you seem to have some Greek influence, though. Paxos... A lot on Greek mythology going on. Is that an interest of yours? |
**Leslie Lamport:** No. It was just -- well, for the Paxos paper it was just something that seemed a good idea at the time. Before I get away from that, I should say that Knuth did all the heavy lifting in building Tex. That was a real remarkable thing that he did. What I did was, in comparison, quite simple... Just bu... |
**Jerod Santo:** I see. So Knuth did all the heavylifting... However, in terms of the tooling, it's still used, it's still prominent. Is it still the majority tool that is used in those communities for specifications and writing? Or is there something that's usurped it since? |
**Leslie Lamport:** No, but I think it's still math and physics... At least it's still the standard -- I don't know about other fields; some fields, it is. Some fields -- I'm sure chemists have never heard of it. |
**Jerod Santo:** Right. Well, as a working software developer, I've definitely heard of it. I don't know it well enough to pronounce it correctly, as you can tell... But I hear it talked about often. Some people love it, other people use it in anger, but everyone seems to use it. It still seems to be the best. And to m... |
**Leslie Lamport:** Oh yeah, that was why Tex needed something on top of it. |
**Jerod Santo:** Okay. |
**Leslie Lamport:** You know, Knuth intended for things to be built on top of it, but I'm not sure he expected so much of the underlying tech to be hidden from the user. By the way, I should say that the whole idea of basically separating the ideas from the typesetting - that was done by Brian Reid in his scribe system... |
\[08:08\] One of my failings in my career is building on things that seem natural, some seem obvious to me, and without realizing that they were new ideas, and that they were significant ideas. I think a prime example of that is in my Time, Clocks paper, where it was inspired by a paper that I read from Paul Johnson an... |
So one of the two significant contributions of that paper was pointing out what this notion of causality - which again, seemed very straight, very obvious to me... But what they had done - they had used timestamps, they added timestamps to the messages. And I just assumed that that was a well known idea, that people di... |
**Jerod Santo:** Ah... So a couple of themes here, the first one being is you building on the work of other people who have come before, and maybe getting in some sense some of the credit on accident. Deflecting, as I can tell... But also, this idea of your ideas, which you've put into papers, and have been read now, a... |
**Leslie Lamport:** Yeah. Because I just expect more of them than it's reasonable to expect of most people. Fortunately, I've had some really smart colleagues who were smarter than me, in fact... And I think if I've been more successful than them, I think it's because of my different way of looking at things, not by be... |
**Jerod Santo:** If we were to focus in on the looking at things differently than other people... Can you describe that, what your angle is, or how you see the world that's eschew, or that's tangential to maybe the way your colleagues were looking at things? What's that different perspective? |
**Leslie Lamport:** Well, it's very simple actually to explain why I'm different from most computer people - it's because I think like a mathematician, and I approach things mathematically, not computationally. |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.