5 ms·

Technical dimensions of programming systems

78 points by jgrodziski 4y ago

phantom32 4y ago
Really interesting writeup!
di4na 4y ago
I find fascinating that the authors analysed only systems that are nearly all "fringe". The Web itself is not really used as described either. It sounds like this work end up producing something that does not even consider the reality of the field but only what the author's imagine the field would be. In that scope it is deeply interesting of course. And what they point to in the opening is important. But i would have loved an implementation that does look at how easy each system make testing. Composability. Refactoring. Linting. Different compilers. Repl. New projects. Compiler driven development. Formatting. Etc There is a wide analysis of the technical side to do there between php and Rust as example. It feels like a missed opportunity.
zabzonk 4y ago
intro: > Programming is done in a stateful environment, by interacting with a system through a graphical user interface. i think all statments in that sentence could be argued with.
- shadowofneptune 4y ago
  All environments are stateful, by necessity. A stateless system would have no input, no output, and consume no electricity. How much the programming requires thinking about that state is one technical dimension.
  - mjburgess 4y ago
    Abstract algorithms given in computer science could meet that condition, so long as they're defined on a turing machine. Eg., you could have a gravitationally stable solar system whose planets, etc. you identified with terms (etc.) and observe that the whole thing runs at zero-energy. This distinguishes abstract algorithms, which are just restatements of mathematics, from "programs" which are algorithms defined for electrical devices with relevant useful capacities.
  - zabzonk 4y ago
    difference between programming environment and program execution environment? if i enter (+ 1 2) which is a lisp program, where is the state? in the context of lisp. but of course, everything is stateful at the end of the day - that's what your cpu is up to.
    - shadowofneptune 4y ago
      Yes, I meant the latter, the actual execution. Kevlin Henney puts it well in his talks (paraphrase): "Functional programming is great! You don't need to worry about the outside environment, it just works! Why, even if you unplug the computer, it just works—no, wait. These are all abstractions." The execution environment is the invariant the webpage starts out with. The programming environment, and its relation to state, can be changed.
  - magicalhippo 4y ago
    A 7400 NAND gate has no state, but it has inputs and outputs and consumes electricity to do useful work. Using lots of them you can build quite non-trivial systems.
alexisread 4y ago
I think they're missing some dimensions. Namely mutability, side effects, formal verification ie. How equational / axiomatic a language is, how static/dynamic a language is (runtime typing etc), how concurrent, how scoped a language is eg. Red vs Koka. All of the above have a massive influence on the structure and intent of a language.
- ahefner 4y ago
  Red vs Koka?
  - alexisread 4y ago
    Sorry, I missed this question: Red (also Rebol) is dynamically scoped: https://stackoverflow.com/questions/21964110/is-there-a-overall-explanation-about-definitional-scoping-in-rebol-and-red https://stackoverflow.com/questions/21964110/is-there-a-over... Koka has to be lexically scoped, to cleanly handle effect types: https://koka-lang.github.io/koka/doc/book.html#sec-hello-world https://koka-lang.github.io/koka/doc/book.html#sec-hello-wor... There is a world of difference between these approaches, both in terms of the runtime/compiler support, and the real-world use cases.
skimdesk 4y ago
An inspiring collection of thoughts. The section on feedback loops had me thinking: Would it be possible to program without feedback (or with minimal feedback). How would that look? Pen and paper still has some feedback as one can see more than can be held in working memory. edit: typo
lincpa 4y ago
[dead]
andromaton 4y ago
The last author is the author of Write your own Excel in 100 lines of F# https://news.ycombinator.com/item?id=20791775 https://news.ycombinator.com/item?id=20791775, 84 comments.

78 points by jgrodziski 4y ago

Really interesting writeup!

I find fascinating that the authors analysed only systems that are nearly all "fringe". The Web itself is not really used as described either. It sounds like this work end up producing something that does not even consider the reality of the field but only what the author's imagine the field would be. In that scope it is deeply interesting of course. And what they point to in the opening is important. But i would have loved an implementation that does look at how easy each system make testing. Composability. Refactoring. Linting. Different compilers. Repl. New projects. Compiler driven development. Formatting. Etc There is a wide analysis of the technical side to do there between php and Rust as example. It feels like a missed opportunity.

zabzonk 4y ago

intro: > Programming is done in a stateful environment, by interacting with a system through a graphical user interface. i think all statments in that sentence could be argued with.

shadowofneptune 4y ago
All environments are stateful, by necessity. A stateless system would have no input, no output, and consume no electricity. How much the programming requires thinking about that state is one technical dimension.
- mjburgess 4y ago
  Abstract algorithms given in computer science could meet that condition, so long as they're defined on a turing machine. Eg., you could have a gravitationally stable solar system whose planets, etc. you identified with terms (etc.) and observe that the whole thing runs at zero-energy. This distinguishes abstract algorithms, which are just restatements of mathematics, from "programs" which are algorithms defined for electrical devices with relevant useful capacities.
- zabzonk 4y ago
  difference between programming environment and program execution environment? if i enter (+ 1 2) which is a lisp program, where is the state? in the context of lisp. but of course, everything is stateful at the end of the day - that's what your cpu is up to.
  - shadowofneptune 4y ago
    Yes, I meant the latter, the actual execution. Kevlin Henney puts it well in his talks (paraphrase): "Functional programming is great! You don't need to worry about the outside environment, it just works! Why, even if you unplug the computer, it just works—no, wait. These are all abstractions." The execution environment is the invariant the webpage starts out with. The programming environment, and its relation to state, can be changed.
- magicalhippo 4y ago
  A 7400 NAND gate has no state, but it has inputs and outputs and consumes electricity to do useful work. Using lots of them you can build quite non-trivial systems.

alexisread 4y ago

I think they're missing some dimensions. Namely mutability, side effects, formal verification ie. How equational / axiomatic a language is, how static/dynamic a language is (runtime typing etc), how concurrent, how scoped a language is eg. Red vs Koka. All of the above have a massive influence on the structure and intent of a language.

ahefner 4y ago
Red vs Koka?
- alexisread 4y ago
  Sorry, I missed this question: Red (also Rebol) is dynamically scoped: https://stackoverflow.com/questions/21964110/is-there-a-overall-explanation-about-definitional-scoping-in-rebol-and-red https://stackoverflow.com/questions/21964110/is-there-a-over... Koka has to be lexically scoped, to cleanly handle effect types: https://koka-lang.github.io/koka/doc/book.html#sec-hello-world https://koka-lang.github.io/koka/doc/book.html#sec-hello-wor... There is a world of difference between these approaches, both in terms of the runtime/compiler support, and the real-world use cases.

skimdesk 4y ago

An inspiring collection of thoughts. The section on feedback loops had me thinking: Would it be possible to program without feedback (or with minimal feedback). How would that look? Pen and paper still has some feedback as one can see more than can be held in working memory. edit: typo

lincpa 4y ago

[dead]

andromaton 4y ago

The last author is the author of Write your own Excel in 100 lines of F# https://news.ycombinator.com/item?id=20791775 https://news.ycombinator.com/item?id=20791775, 84 comments.