A presentation at FunctionalConf in March 2022 in by Brooklyn Zelenka
M A E B e r u t to u f e h t W E N E D A M S A E D I D L O
M A E B e r u t to u f e h t W E N E D A M S A E D I D L O 2022 Edition
Sometimes in order to keep moving forward, not only must you take one step at a time, but you must be willing to look back occasionally and evaluate your past, no matter how painful it is. Looking back lets you know whether or not you are headed in the right direction. –G.K. Adams
Epigram 53 So many good ideas are never heard from again once they embark in a voyage on the semantic gulf –Alan J. Perlis, Epigrams on Programming (1982)
Meta
Brooklyn Zelenka @expede
Brooklyn Zelenka @expede • CTO at Fission (https://fission.codes) • Edge apps (“post-serverless”) • Goal: make back-ends and DevOps obsolete • PLT, VMs, distributed systems • Standards: DIF, UCAN, Ethereum, Multiformats, others • Founded VanFP, VanBEAM • Primary author of Witchcraft, Algae, Exceptional, etc
Meta 🔮
Meta 🔮
Meta 🔮 The BEAM does so much right 👏
Meta 🔮 How do we move forward as an ecosystem, cross language, cross paradigm?
Meta 🔮 How do we move forward as an ecosystem, cross language, cross paradigm?
Meta 🔮 Local Maxima
Meta 🔮 Local Maxima 😄
Meta 🔮 Local Maxima 😍 😄
Meta 🔮 Local Maxima 😍 😄 😭
Meta 🔮 Spoiler Alert 🚨
Meta 🔮 Spoiler Alert 🚨
Meta 🔮 Spoiler Alert 🚨
Meta 🔮 Spoiler Alert 🚨
In the Beginning… 🌴🦖
In the Beginning… 🌴🦖
In the Beginning… 🌴🦖 Epigram 28 Around computers it is difficult to find the correct unit of time to measure progress. Some cathedrals took a century to complete. Can you imagine the grandeur and scope of a program that would take as long? –Alan J. Perlis, Epigrams on Programming (1982)
In the Beginning… 🌴🦖
In the Beginning… 🌴🦖
In the Beginning… 🌴🦖
In the Beginning… 🌴🦖
In the Beginning… 🌴🦖 29 Years
In the Beginning… 🌴🦖 29 Years
In the Beginning… 🌴🦖 29 Years
In the Beginning… 🌴🦖 29 Years
In the Beginning… 🌴🦖 29 Years 36 Years
In the Beginning… 🌴🦖 29 Years 36 Years
Paradigm Redshift 🌈
Paradigm Redshift 🌈
Paradigm Redshift 🌈 It’s really difficult to distinguish a new paradigm from a really bad idea […] The new shiny object is part of the old paradigm – Douglas Crockford, The Power of the Paradigm (2018)
Paradigm Redshift 🌈 Novelty Budget
Paradigm Redshift 🌈 Novelty Budget
Paradigm Redshift 🌈 Novelty Budget
Paradigm Redshift 🌈 Law of Conservation of Complexity
Paradigm Redshift 🌈 Law of Conservation of Complexity Every application has an inherent amount of complexity that cannot be removed or hidden, but only moved from place to place – Larry Tesler
Paradigm Redshift 🌈 The “What If” Tree
Paradigm Redshift 🌈 The “What If” Tree
Paradigm Redshift 🌈 The “What If” Tree
Paradigm Redshift 🌈
Paradigm Redshift 🌈 Let’s start with something alien 🛸👽
Paradigm Redshift 🌈 Something Alien
Paradigm Redshift 🌈 Something Alien {↑1 ⍵∨.∧3 4=+/,̄1 0 1∘.⊖̄1 0 1∘.⌽⊂⍵}
Paradigm Redshift 🌈 Something Alien {↑1 ⍵∨.∧3 4=+/,̄1 0 1∘.⊖̄1 0 1∘.⌽⊂⍵}
Paradigm Redshift 🌈 Something Alien LIFE← {↑1 ⍵∨.∧3 4=+/,̄1 0 1∘.⊖̄1 0 1∘.⌽⊂⍵}
Paradigm Redshift 🌈 Something Alien LIFE← {↑1 ⍵∨.∧3 4=+/,̄1 0 1∘.⊖̄1 0 1∘.⌽⊂⍵} Attribution: wikipedia user LucasVB
Paradigm Redshift 🌈 What can we learn from APL? Cellular Automata & Actors as Organisms 🌸
Paradigm Redshift 🌈 What can we learn from APL? Cellular Automata & Actors as Organisms 🌸
Paradigm Redshift 🌈 What can we learn from APL? Cellular Automata & Actors as Organisms 🌸 • Each step is very simple
Paradigm Redshift 🌈 What can we learn from APL? Cellular Automata & Actors as Organisms 🌸 • Each step is very simple • Reasoning about dynamic organisms is hard!
Paradigm Redshift 🌈 What can we learn from APL? Cellular Automata & Actors as Organisms 🌸 • Each step is very simple • Reasoning about dynamic organisms is hard!
Paradigm Redshift 🌈 What can we learn from APL? Cellular Automata & Actors as Organisms 🌸 • Each step is very simple • Reasoning about dynamic organisms is hard! • Emergent behaviour 😱
Paradigm Redshift 🌈 What can we learn from APL? Cellular Automata & Actors as Organisms 🌸 • Each step is very simple • Reasoning about dynamic organisms is hard! • Emergent behaviour 😱 • VM in your brain to reason at a higher level
Paradigm Redshift 🌈 What can we learn from APL? Cellular Automata & Actors as Organisms 🌸 • Each step is very simple • Reasoning about dynamic organisms is hard! • Emergent behaviour 😱 • VM in your brain to reason at a higher level • We can abstract away some of this • Broadway • Arrows
Paradigm Redshift 🌈 What can we learn from APL? Universal Scaling Law http://www.perfdynamics.com/Manifesto/USLscalability.html
Paradigm Redshift 🌈 What can we learn from APL? Universal Scaling Law Amdahl’s Law http://www.perfdynamics.com/Manifesto/USLscalability.html
Paradigm Redshift 🌈 What can we learn from APL? Universal Scaling Law Amdahl’s Law USL http://www.perfdynamics.com/Manifesto/USLscalability.html
Paradigm Redshift 🌈 What can we learn from APL? Universal Scaling Law Amdahl’s Law Incoherence Penalty http://www.perfdynamics.com/Manifesto/USLscalability.html USL
Paradigm Redshift 🌈 What can we learn from APL? doall / Automatic Parallelism / Cyclic Multithreading 🔁
Paradigm Redshift 🌈 What can we learn from APL? doall / Automatic Parallelism / Cyclic Multithreading 🔁 1. Shared-nothing architecture 2. Good for embarrassingly parallel problems 3. Macro could do a LOT more with this at compile-time 4. Impurity and granular control mean that we don’t get this by default (with good reason)
Paradigm Redshift 🌈 OTP → TOP Table Oriented Programming
Paradigm Redshift 🌈 Table Oriented Programming Naive Tables
Paradigm Redshift 🌈 Table Oriented Programming Naive Tables Name Handle City Brooklyn expede Vancouver Quinn quinnwilton Mountain View Steven icidasset Ghent
Paradigm Redshift 🌈 Table Oriented Programming Naive Tables Name Handle City Brooklyn expede Vancouver Quinn quinnwilton Mountain View Steven icidasset Ghent
Paradigm Redshift 🌈 Table Oriented Programming Naive Tables Name Handle City Brooklyn expede Vancouver Quinn quinnwilton Mountain View Steven icidasset Ghent
Paradigm Redshift 🌈 Table Oriented Programming Perfectly Parallel Control Tables
Paradigm Redshift 🌈 Table Oriented Programming Functions Perfectly Parallel Control Tables Data
Paradigm Redshift 🌈 Table Oriented Programming Functions Perfectly Parallel Control Tables Data
Paradigm Redshift 🌈 Table Oriented Programming Functions Perfectly Parallel Control Tables Data
Paradigm Redshift 🌈 Table Oriented Programming Perfectly Parallel Control Tables Functions Zero dependencies by definition Data
Paradigm Redshift 🌈 Table Oriented Programming The Nth-Dimension 🚀
Paradigm Redshift 🌈 Table Oriented Programming
Composition & Modularity 🧱
Complexity & Modularity 🧱
Complexity & Modularity 🧱 –Alan J. Perlis, Epigrams on Programming (1982)
Complexity & Modularity 🧱 Epigram 6 Symmetry is a complexity-reducing concept; seek it everywhere –Alan J. Perlis, Epigrams on Programming (1982)
Complexity & Modularity 🧱 Epigram 6 Symmetry is a complexity-reducing concept; seek it everywhere Epigram 105 You can’t communicate complexity, only awareness of it –Alan J. Perlis, Epigrams on Programming (1982)
Complexity & Modularity 🧱 What do you mean by “composition”?
Complexity & Modularity 🧱 What do you mean by “composition”?
Complexity & Modularity 🧱 What do you mean by “composition”? HOF Application
Complexity & Modularity 🧱 What do you mean by “composition”? Modularity HOF Application
Complexity & Modularity 🧱 What do you mean by “composition”? Commutativity Modularity HOF Application
Complexity & Modularity 🧱 What do you mean by “composition”? Orthogonality Commutativity Modularity HOF Application
Complexity & Modularity 🧱 What do you mean by “composition”? Focus 🔬
Complexity & Modularity 🧱 What do you mean by “composition”? Focus 🔬 Composition in the data dimension
Complexity & Modularity 🧱 What do you mean by “composition”? Focus 🔬 Composition in the data dimension a b c d e
Complexity & Modularity 🧱 What do you mean by “composition”? Focus 🔬
Complexity & Modularity 🧱 What do you mean by “composition”? Focus 🔬 Composition in the function dimension
Complexity & Modularity 🧱 What do you mean by “composition”? Focus 🔬 Composition in the function dimension f g
Complexity & Modularity 🧱 What do you mean by “composition”? Focus 🔬 Composition in the function dimension f g Da ta flo w h
Complexity & Modularity 🧱 What do you mean by “composition”? Focus 🔬
Complexity & Modularity 🧱 What do you mean by “composition”? Focus 🔬 Composition in the capabilities (protocols and HOFs)
Complexity & Modularity 🧱 What do you mean by “composition”? Focus 🔬 Composition in the capabilities (protocols and HOFs) Data Function
Complexity & Modularity 🧱 What do you mean by “composition”? Execution Symmetry
Complexity & Modularity 🧱 What do you mean by “composition”? Execution Symmetry A program can be developed on a sequential platform, even if it is meant to run on a parallel platform, because the behaviour is not affected by whether we execute it using a sequential or parallel dynamics – Robert Harper, Practical Foundations for Programming Languages (2012)
Complexity & Modularity 🧱 What do you mean by “composition”? Execution Symmetry A program can be developed on a sequential platform, even if it is meant to run on a parallel platform, because the behaviour is not affected by whether we execute it using a sequential or parallel dynamics – Robert Harper, Practical Foundations for Programming Languages (2012) t
Complexity & Modularity 🧱 What do you mean by “composition”? Execution Symmetry A program can be developed on a sequential platform, even if it is meant to run on a parallel platform, because the behaviour is not affected by whether we execute it using a sequential or parallel dynamics – Robert Harper, Practical Foundations for Programming Languages (2012) t
Complexity & Modularity 🧱 What do you mean by “composition”? Execution Symmetry A program can be developed on a sequential platform, even if it is meant to run on a parallel platform, because the behaviour is not affected by whether we execute it using a sequential or parallel dynamics – Robert Harper, Practical Foundations for Programming Languages (2012) t
Complexity & Modularity 🧱 What do you mean by “composition”? Execution Symmetry A program can be developed on a sequential platform, even if it is meant to run on a parallel platform, because the behaviour is not affected by whether we execute it using a sequential or parallel dynamics – Robert Harper, Practical Foundations for Programming Languages (2012) t
Complexity & Modularity 🧱 What do you mean by “composition”? Execution Symmetry A program can be developed on a sequential platform, even if it is meant to run on a parallel platform, because the behaviour is not affected by whether we execute it using a sequential or parallel dynamics – Robert Harper, Practical Foundations for Programming Languages (2012) t
Complexity & Modularity 🧱 What do you mean by “composition”? Execution Symmetry A program can be developed on a sequential platform, even if it is meant to run on a parallel platform, because the behaviour is not affected by whether we execute it using a sequential or parallel dynamics – Robert Harper, Practical Foundations for Programming Languages (2012) t 👍 ! s e t u m m o c It
Complexity & Modularity 🧱 What do you mean by “composition”? Execution Symmetry A program can be developed on a sequential platform, even if it is meant to run on a parallel platform, because the behaviour is not affected by whether we execute it using a sequential or parallel dynamics – Robert Harper, Practical Foundations for Programming Languages (2012) t 👍 ! s e t u m m o c It
Complexity & Modularity 🧱 What do you mean by “composition”? Execution Symmetry A program can be developed on a sequential platform, even if it is meant to run on a parallel platform, because the behaviour is not affected by whether we execute it using a sequential or parallel dynamics – Robert Harper, Practical Foundations for Programming Languages (2012) t 👍 ! s e t u m m o c It
Complexity & Modularity 🧱 What do you mean by “composition”? Explicit Data Flow
Complexity & Modularity 🧱 What do you mean by “composition”? Explicit Data Flow
Complexity & Modularity 🧱 What do you mean by “composition”? Explicit Data Flow
Complexity & Modularity 🧱 What do you mean by “composition”? Explicit Data Flow
Complexity & Modularity 🧱 How Modular are Modules?
Complexity & Modularity 🧱 How Modular are Modules? (and libraries)
Complexity & Modularity 🧱 How Modular are Modules? (and libraries)
Complexity & Modularity 🧱 How Modular are Modules? Module-Level Modularity
Complexity & Modularity 🧱 How Modular are Modules? Module-Level Modularity “Hot-swappable dependencies”
Complexity & Modularity 🧱 How Modular are Modules? Module-Level Modularity “Hot-swappable dependencies”
Complexity & Modularity 🧱 How Modular are Modules? Module-Level Modularity “Hot-swappable dependencies”
Complexity & Modularity 🧱 How Modular are Modules? Module-Level Modularity “Hot-swappable dependencies”
Complexity & Modularity 🧱 How Modular are Modules? Hacking Extending the Module System
Complexity & Modularity 🧱 How Modular are Modules? Hacking Extending the Module System
Complexity & Modularity 🧱 How Modular are Modules? Hacking Extending the Module System
Complexity & Modularity 🧱 How Modular are Modules? Higher Order Modules 💖 Behaviours
Declarative Embedded DSLs 🍏🍊
Declarative Embedded DSLs 🍏🍊
Declarative Embedded DSLs 🍏🍊 I have regarded it as the highest goal of programming language design to enable good ideas to be elegantly expressed – Tony Hoare, Turing Aware Lecture, 1980
Declarative Embedded DSLs 🍏🍊 Isn’t Everything a DSL?
Declarative Embedded DSLs 🍏🍊 Isn’t Everything a DSL? Wide Narrow
Declarative Embedded DSLs 🍏🍊 Isn’t Everything a DSL? m g i d ra a P Wide t a P n r te n a L a u g e g D L S k n r o o i e t w c a e a c f i l r m p e a t r p n F I A Narrow
Declarative Embedded DSLs 🍏🍊 Isn’t Everything a DSL? Counterexample
Declarative Embedded DSLs 🍏🍊 Isn’t Everything a DSL? Counterexample This is just Elixir!
Declarative Embedded DSLs 🍏🍊 Isn’t Everything a DSL? Algebraic Data Type eDSL
Declarative Embedded DSLs 🍏🍊 Isn’t Everything a DSL? Algebraic Data Type eDSL
Declarative Embedded DSLs 🍏🍊 Isn’t Everything a DSL? Algebraic Data Type eDSL
Declarative Embedded DSLs 🍏🍊 Isn’t Everything a DSL? Algebraic Data Type eDSL
Declarative Embedded DSLs 🍏🍊 Business Language for Business Time 💼
Declarative Embedded DSLs 🍏🍊 Business Language 💼
Declarative Embedded DSLs 🍏🍊 Business Language 💼 What can we learn from COBOL? (Yes, really. COBOL.) Toggl, How to Kill at the Dragon in 9 Programming Languages
Declarative Embedded DSLs 🍏🍊 What Can We Learn from COBOL?
Declarative Embedded DSLs 🍏🍊 What Can We Learn from COBOL? When No One Wants to Go To Jail 👩⚖
Declarative Embedded DSLs 🍏🍊 What Can We Learn from COBOL? When No One Wants to Go To Jail 👩⚖ • Needs to be readable by lawyers • (Who can’t read code) • Formal methods & static analysis • Down to the compiler, of course • An unholy union of COBOL and Prolog
Declarative Embedded DSLs 🍏🍊 What Can We Learn from COBOL? When No One Wants to Go To Jail 👩⚖ • Needs to be readable by lawyers • (Who can’t read code) • Formal methods & static analysis • Down to the compiler, of course • An unholy union of COBOL and Prolog
Declarative Embedded DSLs 🍏🍊 Upside
Declarative Embedded DSLs 🍏🍊 Upside
Declarative Embedded DSLs 🍏🍊 But We Have a Problem
Declarative Embedded DSLs 🍏🍊 But We Have a Problem (Hint: it’s inflexibility)
Declarative Embedded DSLs 🍏🍊 🏖 Shallow Embedding 🦀
Declarative Embedded DSLs 🍏🍊 🏖 Shallow Embedding 🦀 i.e. The “Quick and Dirty” Way
Declarative Embedded DSLs 🍏🍊 Shallow Embedding
Declarative Embedded DSLs 🍏🍊 Shallow Embedding • Just use the built-in AST • What it can represent is limited • e.g. Ecto, Algae
Declarative Embedded DSLs 🍏🍊 Shallow Embedding • Just use the built-in AST • What it can represent is limited • e.g. Ecto, Algae
Declarative Embedded DSLs 🍏🍊 Shallow Embedding • Just use the built-in AST • What it can represent is limited • e.g. Ecto, Algae
Declarative Embedded DSLs 🍏🍊 Shallow Embedding • Just use the built-in AST • What it can represent is limited • e.g. Ecto, Algae
Declarative Embedded DSLs 🍏🍊 🌊 Deep Embedding 🦈
Declarative Embedded DSLs 🍏🍊 🌊 Deep Embedding 🦈 “Better AST”
Declarative Embedded DSLs 🍏🍊 Deep Embedding Three Steps
Declarative Embedded DSLs 🍏🍊 Deep Embedding Three Steps
Declarative Embedded DSLs 🍏🍊 Deep Embedding “Build a Game Plan” Example
Declarative Embedded DSLs 🍏🍊 Deep Embedding “Build a Game Plan” Example
Declarative Embedded DSLs 🍏🍊 Deep Embedding Tradeoffs
Declarative Embedded DSLs 🍏🍊 Deep Embedding Tradeoffs •More work to write (write your own AST) •Way more powerful (full control) •Precisely the vocabulary that you need — exact surface area •Can check more things about the meaning of your code •Logic-as-data is MUCH simpler to debug than running functions •Time travelling debugging! •Unlike protocols, you’re not locked into one canonical implementation
Declarative Embedded DSLs 🍏🍊 Deep Embedding Combine Powerful, Modular, Reusable DSLs!
Declarative Embedded DSLs 🍏🍊 Deep Embedding Desugar
Declarative Embedded DSLs 🍏🍊 Deep Embedding “Huh, this kinda feels like GenServer” 😉
Declarative Embedded DSLs 🍏🍊 Deep Embedding “Huh, this kinda feels like GenServer” 😉
Declarative Embedded DSLs 🍏🍊 Deep Embedding “Huh, this kinda feels like GenServer” 😉
Declarative Embedded DSLs 🍏🍊 Deep Embedding “Huh, this kinda feels like GenServer” 😉
Declarative Embedded DSLs 🍏🍊 Deep Embedding With Their Powers Combined!
Declarative Embedded DSLs 🍏🍊 Deep Embedding One Last Line
Declarative Embedded DSLs 🍏🍊 Deep Embedding One Last Line
Declarative Embedded DSLs 🍏🍊 Deep Embedding More Flexible Than protocols
Declarative Embedded DSLs 🍏🍊 Deep Embedding More Flexible Than protocols
Let’s Make New Mistakes! 💥✨
Let’s Make New Mistakes! 💥✨ 5 Problems for the Next 30 Years of BEAM
Let’s Make New Mistakes 5 Problems for the Next 30 Years of BEAM
Let’s Make New Mistakes 5 Problems for the Next 30 Years of BEAM
Let’s Make New Mistakes 5 Problems for the Next 30 Years of BEAM
Let’s Make New Mistakes 5 Problems for the Next 30 Years of BEAM
Let’s Make New Mistakes 5 Problems for the Next 30 Years of BEAM
Let’s Make New Mistakes 5 Problems for the Next 30 Years of BEAM
Parting Thought 🧠
–Alan J. Perlis, Epigrams on Programming (1982)
Epigram 101 –Alan J. Perlis, Epigrams on Programming (1982)
Epigram 101 Dealing with failure is easy: –Alan J. Perlis, Epigrams on Programming (1982)
Epigram 101 Dealing with failure is easy: work hard to improve. –Alan J. Perlis, Epigrams on Programming (1982)
Epigram 101 Dealing with failure is easy: work hard to improve. Success is also easy to handle: –Alan J. Perlis, Epigrams on Programming (1982)
Epigram 101 Dealing with failure is easy: work hard to improve. Success is also easy to handle: you’ve solved the wrong problem. –Alan J. Perlis, Epigrams on Programming (1982)
Epigram 101 Dealing with failure is easy: work hard to improve. Success is also easy to handle: you’ve solved the wrong problem. Work hard to improve. –Alan J. Perlis, Epigrams on Programming (1982)
🎉 Thank You, India 🇮🇳 brooklyn@fission.codes https://fission.codes github.com/expede @expede
