Introduction

Why such a strange course?

Where I'm coming from?

  • PhD, University of Cambridge
    Context-aware programming languages
  • Microsoft Research Cambridge
    F# and applied functional programming
  • The Alan Turing Institute, London
    Expert and non-expert tools for data science
  • University of Kent, Canterbury
    History and programming systems

Demo

Coeffects playground

Did this to get my PhD...

How to show potential uses of theoretical work?

Tiny type system running in the web browser

Tiny demos of two potential applications

Programming Languages

Programming is
writing code

Formal semantics, implementation, paradigms, types

We know how
to study this!

Programming Systems

Interacting with a stateful system

Feedback, liveness, interactive user interfaces

But how do we
study this?

Demo

The Gamma project

Making programmatic data exploration accessible to non-programmers

From language to system

Small typed language

Interaction is the key.
This is why it works!

Paradigm shift in 1990s

From systems to languages

  • From running system to code
  • From state & interaction to semantics
  • Incommensurable ways of thinking!

History of science matters!

  • How did we get where we are?
  • What ideas got lost along the way?
  • How to recover them?

Research

What do I work on today?

  • History and philosophy of computing
  • Programming languages, types and theory
  • Interactive programming environments
  • Will artificial intelligence make me obsolete?

Programming languages at D3S

Growing group of great people

  • Jan Vitek (via Northeastern)
  • Aleksander Boruch-Gruszecki
  • Also talking to PRL-PRG at CTU!

Growing number of activities!

  • Programming languages reading group
  • New courses (NSWI182, NPRG075, NPRG077)
  • PL topics at the regular D3S seminar

Starting points

Writing tiny systems

Two uses of tiny systems

Education

Best way to learn?
Write it on your own!

Understand principles
As well as subtle details

I hope you'll have fun!
Doing more with less?

Research

Imagine new paradigms
Variable names

Focus on interaction
How exactly did it work

Ignore practical details
New mode of interaction

Teaching tiny systems

(Kamin, 1990)

Used in multiple
courses worldwide

Examples in Pascal

Languages covered are APL, Clu, LISP, Prolog, Smalltalk, Scheme, SASL

Not always focused
on the key aspect

Tiny systems and AI

(Schank, Riesbeck, 1981)

Miniature implementations of 5 Yale AI lab programs

Faster, more efficient, easier to understand, modify and extend

"Miniatures, demos and artworks" by Warren Sack

Tiny systems and ML

(Distill, 2016-2021)

Five affordances of interactive articles

Connecting people & data
Making systems playful
Prompting self-reflection
Personalizing reading
Reducing cognitive load

Programming models

Learning by implementing

Programming models

Language paradigms

  • Functional programming
  • Imperative programming
  • Object-oriented programming
  • Logic programming

Programming models

System interaction

  • Image-based programming model
    Programming system is always running
  • Interactive and live programming
    System provides continuous feedback
  • Incremental or reactive evaluation
    Recompute on edit or when new data come

Demo

Logic programming in Prolog

Demo

Object-orientation in Smalltalk

What really matters?

Static structure

  • Source code of the program
  • What you have at the start

Dynamic structure

  • Runtime data structures
  • What else do you need to run

Logic of evaluation

  • How the dynamic state evolves?
(* A term like 'father(william, X)'
   consists of predicate 'father',
   atom 'william' and variable 'X' *)
type Term =
  | Atom of string
  | Variable of string
  | Predicate of string * Term list

(* A rule 'head(...) :- body.' *)
type Rule =
  { Head : Term
    Body : Term list }

(* A program is a list of rules *)
type Program = Rule list

Why interpreters?

A good way to explain the structures!

Functional data types for the static and dynamic structure

A function to model the evaluation logic

Operational semantics

Standard approach
to programming language theory

Equations vs. Code

Code actually runs!
Easier to write?

Course scope

What is not covered?

  • Syntax choices and writing parsers
  • Compilation and JIT-based runtimes
  • Formal semantics and correctness
  • Supporting real-world use cases

Tiny systems

Programming systems research

Academic research

What are we trying to study?

  • Basic essential principles
  • In isolation from other factors
  • You have to ignore a lot!

What to ignore in programming?

  • Efficient implementation?
  • Wide-spread user adoption?
  • User interface of editor tools?

Programming language theory

Ignore implementation and practical features

Prove that the core idea is formally sound

Human-computer interaction (HCI)

Ignore inner working and implementation

Show that users can actually use it and how

Performance evaluation

Ignore usability and design implications

Show that you can do better than a baseline

Tiny systems

What can we study?

  • Can talk about stateful interactive systems
  • Implement key aspects of inner working
  • Reconstruct interesting past systems
  • But cannot be printed on 12 pages of A4

Demo

C64 BASIC

Why study universally disliked programming language?

Somehow allowed everyone to program!

Interesting mode of interaction!

Course background

Getting started with F#

The F# programming language

What is F# about?

  • Functional-first based on OCaml
  • Great interop with .NET and JS
  • Open-source (MIT) with team in Prague!

Who uses F# for what?

  • Consultancies for full-stack web dev
  • Finance and insurance companies for modelling
  • TU Kaiserslautern for systems biology
  • Success stories like Jet.com

Why F#?

Building tiny programming systems

  • Algebraic data types for structure modelling
  • Mostly functional is great for logic
  • Runs everywhere & has nice tools
  • I like the language and can help you!

Demo

First look at F#

Elmish architecture

Functional interactive user interface development

Types for application
State and user Event

Functions to render
and update state

Demo

Building a TODO list in F#

Closing

Write your own tiny system

Practical details

Course structure

  • Videos + bi-weekly hands-on labs
  • Watch before & finish after!
  • Remote possible - email me
  • Check the schedule on course web site!

To get the credits

  • Active participation in the labs
  • Awarded based on a git repo
  • Complete basic tasks for 4/6 systems

Conclusions

Write your own tiny programming system(s)!

  • Learn interesting programming models!
  • Nice programming research methodology
  • We have projects and PhD positions available :-)

Tomáš Petříček, 309 (3rd floor)
petricek@d3s.mff.cuni.cz
https://tomasp.net | @tomaspetricek
https://d3s.mff.cuni.cz/teaching/nprg077

References