from Hacker News

Ask HN: How would a programming language look if designed by non-programmer

by tester34 on 12/17/21, 11:15 AM with 130 comments

Since it's hard to find such a person, who understands CS/Math/can program machines but never used "normal" programming languages

I do wonder how it'd be designed, maybe current approach sucks?

OOP is beautiful itself, it *enables normal people* to model complex systems that'll be running on computers, but can we get even better?

  • by danpalmer on 12/17/21, 11:43 AM

    It would probably look a lot like Excel.

    Most people I work with who are non-programmers but who could benefit from coding spend a lot of their working life in Excel. When we teach them SQL and give them access to our data warehouse, a lot of the time the explanations that stick are thinking of it like a big spreadsheet with lots of pages and columns/rows. Spreadsheets are fundamentally very visual, and have few abstractions to get your head around before you can start work. (There have been "visual programming languages", but these often look like graph editors and I think that's a complex enough abstraction that it complicates the process).

    From this we've then seen a whole load of tools that are trying to be "smarter" spreadsheets. Things like Airtable. Is it a spreadsheet? Is it a database? Is it a no-code application? Is it a project management system? Is it a calendar? Well, yes. Retool, and others also fit into this. They all have a lot of spreadsheet like functionality.

  • by j-pb on 12/17/21, 11:38 AM

    50 years ago, these people were quite common. There were only mathematicians and engineers, but programmers didn't exist yet.

    The good: Excel, Dataflow Diagrams (commonly found in game, music and arts), SQL, Declarative Dataflow Paradigms in general. Mutablity is counterintuitively counterintuitive if you don't have a model of the underlying execution system.

    The weird: Prolog, APL, languages done by mathematicians and logicians

    The ugly: SPSS, R, two languages done by statisticians, and they are absolutely abhorrent, and often missused.

    If we reinvented programming today my guess is that it would be a lot less text driven and a lot more data oriented.

  • by jasonhong on 12/17/21, 2:39 PM

    You will be very interested in my colleague Brad Myers' work on Natural Programming. http://www.cs.cmu.edu/~NatProg/

    In particular, one of his PhD students John Pane did a number of studies looking at how people would naturally describe the behaviors in Pac Man. It was really nice work.

    You should also take a look at Programming by Demonstration. One book, Watch What I Do, is published online in its entirety by the editor Allen Cypher, see here http://acypher.com/wwid/.

    Another book in this space, Your Wish is My Command, is also published online https://web.media.mit.edu/~lieber/Your-Wish/

    Both of these are collections of research papers showing how people can demonstrate what they want to a computer, and have the computer figure out what they want. There's a more recent book on programming by demonstration, but I don't recall the title.

    But definitely check out the first link on Natural Programming, Brad Myers is one of the world experts on this topic.

  • by depaya on 12/17/21, 3:06 PM

    Take a look at Ladder Logic: https://en.wikipedia.org/wiki/Ladder_logic

    I don't think this was necessarily created by non-programmers, but it was certainly created for them. The language is designed to mimic electrical schematics with the idea being electricians, electrical engineers, etc can easily program and debug in the language.

    I'm a software engineer who has worked extensively programming PLCs with Ladder Logic and I must say it can be quite fun to work with.

  • by OJFord on 12/17/21, 1:06 PM

    AppleScript is sort of an attempt at that isn't it?

    I got there by thinking 'probably not that far off python, but even more pseudocodey, a load of redundant keywords like `for every item in` instead of `for v in`' -- and such constructs probably wouldn't bind variables, you'd repeat yourself referring to `the item` or something.

    I also think there'd probably be a lot of focus on (high school) mathematical functions (since that would be the expectation of what computers must be like) and graphics (because that's what computers look like). So everything would be about identifying pixels on the screen and colouring them, locating the cursor by position, that sort of thing.

  • by JackMorgan on 12/17/21, 1:22 PM

    I'm not sure I'd agree that OOP is beautiful. It's just nouns with verbs inside, which doesn't seem particularly better than nouns with verbs outside.

    I definitely would not agree that it "enables normal people to model complex systems". Not once in my entire career have I seen a business person provide anything of any value to a software's architecture. Even thinking they could is so strange to me I don't even know where to start to unpack it. Software architecture is so different to the processes it models, it would be nonsensical to even have them try. Like asking someone illiterate to organize a book.

    Wasn't that whole concept just marketing by Sun to sell Java and OOP to the world? These days everyone using Java uses so much spring and xml I'm not sure it even makes sense to call it OOP anymore. It's more like AOP+DI with a little OOP thrown in.

    If anything, I'd think trying to have a customer design your objects would result in a tremendous mess. The nouns and verbs they would think of are going to be extremely limiting once you add in the common needs like caching, parallelism, network boundaries, async processing, auth, logging, monitoring, protocol formatting, API versioning, etc. The hard parts we need help on from business people are understanding the business models completely, not where X function should live and if it should be in a class or not. And even if they did have strong opinions about that, logic is actually much easier to test when it's not in OOP, and I'd rather have automated unit tests.

  • by open-source-ux on 12/17/21, 12:10 PM

    One possibility: the non-programmer is more open to designing a language that is more than just text.

    Programming in plain text (in a monospace font) hasn't changed for decades. It doesn't look like it will change for the foreseeable future either. This isn't because plain text is inherently superior. We're still relying on clumsy, cryptic syntax determined by the limited characters available on a keyboard.

    The simplicity of text is very appealing to developers. But arguably, it's modern IDEs that make dealing with plain text tolerable - without them, the simplicity of text would rapidly lose its shine. And modern IDEs are enormous in their complexity and millions of lines of code.

    So new programming ideas may come from unexpected quarters. And maybe less likely from developers conditioned by existing languages and tools.

  • by AnimalMuppet on 12/18/21, 4:21 AM

    I worked with an artist to design user interface menus for a video graphics system. I'd design a menu that did what was needed. It was ugly. The artist would look at it, and say "That's like picking your nose with boxing gloves on."

    Then he'd draw this beautiful menu. And I'd say, "How does it know that you want to do X instead of Y?"

    We'd go back and forth, and wind up with something that was both graphically good and also unambiguous.

    If you just have the non-programmer doing the design, I worry that it would stop at the "ambiguous" stage. There's a level of technical rigor that is needed, and non-programmers may find it hard to achieve it.

  • by bsenftner on 12/17/21, 1:01 PM

    Perhaps I'm grumpy, but questions like this irk me. A programming language designed by a non-programmer would simply not work. It would have non-logical magic the non-programmer expects us programmers to create, things like the "do what I mean" function. One would have to seek out someone with a nearly programming language formality in their logical thinking, such as a mathematician, or linguist. Whatever such a non-programmer would create, it would have giant logic holes they never imagined, rendering the toy "language" unusable for anything beyond the toy concepts the non-programming imagined, and those would require that "do what I mean" function to work at all.
  • by efitz on 12/17/21, 1:02 PM

    It would look like HyperTalk. I still get a kick out of a programming language that feels like English, prepositions and all.
  • by ksec on 12/17/21, 3:17 PM

    It will be more like Pascal and Ada than something like C or Rust.

    I think Non-programmers would want programming to be more graphical, Diagram Dataflow and Graph-like with the help of "script" / functions describing actions. Something like the how current Game Engine are structured.

    May be Game Engine should be programming engine.

  • by Jtsummers on 12/17/21, 3:52 PM

    Ad hoc, it would be The Homer of programming languages if allowed to grow unfettered. Short of that, it would be a DSL oriented towards their particular needs. Possibly graphically oriented. Meaning something more like LabView or perhaps Excel, with a heavy visual component that lends itself well to people who are not experienced enough to visualize a complex abstract system or who need to communicate the graphical model to others.

    For instance, a friend worked in "manpower" (operations research work, trying to understand and improve the workflow of a 15k person industrial workforce). One of the tools was a hybrid of a graphical model + R-kin language. The model looked like any ordinary flowcharting kind of thing, though the data attached were numbers like throughput, reliability, etc. And when "executed" it would produce a report based on real or modeled data about expected results. The graphical model lended itself well both for the users (would you want to write out thousands of lines of: engine removed -> load engine on cart -> move cart to workroom -> clean engine -> ...?) and the people it was meant to communicate with ("Is this really your workflow? Do we have a detail wrong?"). The R-kin language was something that she made use of, but mostly for generating reports and it was, to her, "math" not "programming".

  • by Smithalicious on 12/17/21, 12:59 PM

    Version 1 might look odd but version 3 would look an awful lot like a normal programming language. There is no special "programmer" phenotype of person, there's only people who program. All of our current programming languages ultimately trace their history to non-programmers.
  • by saltyfamiliar on 12/17/21, 10:16 PM

    One thing I've noticed always seems to amaze non-programmers as they're taking their first foray into using computers in more advanced ways (programming, scripting, even just using a command-line) is how little room there is for typos. I've frequently heard things like, "if you forget a single semicolon the whole thing doesn't work!" To me, it just seems obvious that if you make a mistake, the computer won't know what to do with your code. But maybe it doesn't really have to be that way.

    So my guess is that a programming language created by a non-programmer would be extremely tolerant to typos. References to non-existent variables and functions would resolve to the nearest existing options. Semicolons, commas, parentheses and brackets would be strictly optional. This seems to be closer to the way people naturally use language.

  • by knbknb on 12/17/21, 12:28 PM

    TYPOScript, a Domain-Specific Language (DSL) for customizing the TYPO3 Content Management System. Typoscript is not a turing-complete language.

    For starters, TYPOscript has very odd semantics for terms like "Object", "datatype", "register", and so on. You can have lines of code like "page = ", or "page < ".

    Typoscipt was designed in the late 1990s, then by an autodidact or very young CS student(s). I don't know exactly, and I don't want to blame anyone. (nothing personal).

    However, as a consequence, as a TYPO3 user I have countless files in my work-related directories that have a ".ts" filename extension - now used by TypeScript a well designed language for which rich toolsets exist.

    So when I open a TYPOscript .ts file, practically every word on every line has these red squiggly lines from the linter ... very scary.

    TYPOscript has its strengths, though.

  • by knbknb on 12/17/21, 12:47 PM

    Principia

    A work from the early 20th century, mathematics, logic.

    I think some expert called Bertrand Russell's and A.N. Whitehead's "Principia Mathematica" initiative a "bizarre" piece of work, when seen from the perspective of a programming language designer.

    I can't make a qualified statement about this, as I am neither a mathematician nor a language designer. And I cannot find the exact quote on the internet, sorry. Just saying.

    In code? See for yourself :

    https://www.principiarewrite.com/ or https://github.com/LogicalAtomist/principia/blob/master/PL.v

  • by jerf on 12/17/21, 3:26 PM

    It's very hard to provide any interesting answer to that. I know programmers like to be all egalitarian and stuff, but...

    What if we ignored all the road workers and asked non-construction people who kinda know how to build some stuff in their house to design & build a highway instead?

    What if ignored all the architects and asked a non-architect who has put up a couple of sheds from scratch to build the next skyscraper?

    What if we ignored all the doctors and asked a random person who knows some biology to treat cancer?

    I think this may make it more obvious how this would go.

    You may, very rarely, get some sort of interesting insight out of such things. I won't claim the probability is a flat mathematical zero. But it's not a very rich vein of insights. There's much richer veins available.

  • by 32gbsd on 12/27/21, 11:45 PM

    OOP is trash. Its the patadox you face in which "normal" people are forced to think in a box that limits their way of thinking about abstract concepts. Everything is not an object. Once you remove oop you will be able model much more.
  • by pmelendez on 12/17/21, 3:16 PM

    Perl comes to mind, given the fact that was trained as a linguistic: https://en.wikipedia.org/wiki/Larry_Wall
  • by exogenousdata on 12/17/21, 3:32 PM

    I believe this aphorism applies to programming languages, "everything should be as simple as it can be, but not simpler". And at the risk of not answering the question, the reality is that no one language would ever be sufficient because a programming language is a tool, not a solution. And different jobs/goals require different tools. Therefore, to paraphrase from Malcom Gladwell, there is no such thing as the perfect programming language for non-programmers, only the perfect programming languages.

    From there I'd imagine the best thing we can do is discuss features/tools that are used today that would be good for non-programmers. As many commenters have suggested, this will probably appear like the existing suite of tools that non-programmers use.

    First, maybe an interactive graphical component that allows a users to define the flow of data. Something like Visio to allow users to draw data & input/output, mixed with a component like MSAccess to define the data model. Then a natural language tool that would take normal language and apply rules on the data (Hypercard). Finally, a visual component (like Excel) to validate that the data is correct and potentially generate reports for other non-programmers.

  • by tlb on 12/17/21, 11:31 AM

    It'd be a fascinating experiment. The languages we use all have a great deal of shared history.

    Essentially all modern languages were designed by people who were good enough programmers to implement compilers/interpreters/runtimes in a language like C, so it's hard to avoid some intellectual crossover. You'd need to have a separate team implementing the design proposed by this outsider.

  • by Arete314159 on 12/17/21, 9:29 PM

    A programming language for "normal" people would not start counting at 0. For starters.
  • by kbob on 12/17/21, 11:33 AM

    Go look at some of the programming languages from many years ago, when most of the good stuff we like hadn't been thought of.

    A good starting point is the video, 50 in 50, by Guy Steele and Richard Gabriel. It reviews 50 programming languages from the first 50 years of computing.

  • by charlieyu1 on 12/17/21, 3:21 PM

    Either it is abondoned in the middle, or it would become something like it were developed by a programmer.

    People thought writing code is the hard part. It is not. It is the underlying problem that is hard. And no language can really help you about that

  • by tyingq on 12/17/21, 2:45 PM

    Early languages might be good examples, since there wasn't really a defined path of programmer. COBOL, for example, has the feel of a smart, but not traditional programmer approach. Grace Hopper was a mathematics PhD.
  • by tiborsaas on 12/17/21, 12:30 PM

    It would look like GPT-3, you enter text and the system translates that to code machines can understand.

    A programming language will never be a solution for people who can't program, break down problems to smaller chunks and be able to generalize and create abstractions.

    Think about someone who can't drive. To them no matter how good a car is, a car alone will never be the solution to move around (ignore self driving cars for now since you are still required a driving license).

    For these people one solution might be a taxi driver who can translate commands to actions to move the car to the destination.

  • by hzlatar on 12/17/21, 6:58 PM

    I have an anecdote. Back then when I was studying CS on the exam for computer architecture class, we got a problem to solve: "Define the instruction set for your imaginary CPU and then use it to write a program for managing an elevator."

    They expected us to imagine a simple instruction set with 7-10 commands. But one clever guy wrote this: "My CPU has only one instruction - manage_elevator. The code for elevator management is: "manage_elevator". He passed with A.

    I guess that is how a non-programmer would imagine a programming language.

  • by logicalmonster on 12/17/21, 4:49 PM

    Visual programming approaches are fascinating, but something like Applescript with its syntax that resembles natural language is probably the most realistic attempt at something non-programmer friendly.

    This is more of a fantasy answer, but I think Star Trek: TNG characters describing a desired holodeck scene to the ship's computer is probably an example of what the ultimate programming language would look like. No memorization of syntax required: computers are just magic tools that always understand what you intend.

  • by 999900000999 on 12/17/21, 3:45 PM

    Python is pretty close.

    I would pay good money to watch random people try to do something like get all the names from a CSV file in different programing languages.

    Have them on a zoom with a Senior Software Engineer. But said Senior Engineer is not allowed to write any code directly.

    In this hypothetical game show, I'd imagine most software engineers would pick python to try and teach remotely. Although it would be funny to see the same exercise with Rust.

  • by thom on 12/17/21, 12:42 PM

    Perhaps like Inform:

    http://inform7.com/

    Although I suppose in the age of GPT-3 that will probably take over.

  • by cranium on 12/17/21, 12:46 PM

    Coding is organizing and modifying information. If you didn't code before, you cannot comprehend all the subtleties and will end up squeezing new concepts in your language as you start understanding them. Classical recipe for delicious spaghetti code and we all started there!

    Through iterations and deliberate design choices, you may end up with a workable language. But you are no longer a non-programmer by then.

  • by vb6sp6 on 12/17/21, 6:05 PM

    It would look like english (or whatever language you speak)

    "show me orders placed today in Excel"

    "send an email to bob@company.com with the order details"

    "backup the database to my dropbox account and do it every monday"

    "setup a server in new york"

    Non-technical people do this all day, giving these simple instructions to the programmers and letting them figure out the details :)

  • by skadamat on 12/21/21, 11:22 PM

    Required viewing is Bret Victor's excellent talk "Humane Representation of Thought":

    https://vimeo.com/115154289

  • by 8K832d7tNmiQ on 12/17/21, 7:34 PM

    I've seen lots of journal papers in my field that use labVIEW to write an application that uses a node-based model or data-flow coding, similar to how Blender handles Shader nodes.

    In general, it helps students to write a quick application and focus to the important part at the cost of limiting the environment to labVIEW.

  • by t-3 on 12/17/21, 2:09 PM

    Well, I'm not a programmer/have never worked with computers for a living and have no formal education in computing, and I have recently been designing my own language as a learning/fun exercise. It looks a lot like forth and ksh had a baby that grew up to become a state machine.
  • by petra on 12/17/21, 12:34 PM

    Languages are not usually aimed at no-programmers. But low-code tools are aimed at them.

    They're done by well funded companies, so I assume a lot of work(by very skilled people) has be done at understanding the answer to OP's question, and so a lot could be learned from them.

  • by NinjaViking on 12/17/21, 11:46 AM

    COBOL was supposed to be just that. It is not considered a nice or beautiful language generally.
  • by thehappypm on 12/17/21, 2:23 PM

    LabView or a similar visual GUI would be my guess. I always found it to be pretty intuitive.
  • by jcun4128 on 12/17/21, 11:36 AM

    Just commenting: before I really I got into web dev, I dropped some time on this Microsoft platform called Touch Develop. It was interesting but I was glad I did not get locked into that way of thinking.

    I could see the appeal of something like voice and copilot.

  • by onion2k on 12/17/21, 12:40 PM

    Perhaps something like the node editor in game engines - a visual system for connecting inputs to outputs and transforming the data as it goes through the graph. Non-technical users seem understand the analogy very intuitively.
  • by kangnkodos on 12/17/21, 3:27 PM

    It would look like the Rockstar language. https://www.youtube.com/watch?v=gdSlcxxYAA8&t=2959s
  • by mikewarot on 12/17/21, 4:10 PM

    Actually, it wouldn't be hard to find out, it would just take some funding, or someone who can write a compiler/repl/IDE/wherever volunteering time to follow the direction of a layman.
  • by tmaly on 12/17/21, 5:43 PM

    I could imagine a declarative style language like Prolog could fit this.
  • by mikewarot on 12/17/21, 10:52 PM

    One thing to do would be Rich source text, with embedded graphics, video, etc. directly in the source, instead of external files.

    You could also support inline literal values, etc. similar to Color Forth.

  • by wodenokoto on 12/17/21, 8:52 PM

    My lazy response:

    If designed by a well educated philosopher or linguist, probably similar to some 70s or 80s language.

    If by someone with no prerequisites then it would be full of ambiguity - just like natural languages.

  • by TT-392 on 12/18/21, 11:00 AM

    So, not made by non programmers, but definitely made for non programmers. The blender node system, it is a graphical programming language that is actually very usable.
  • by charcircuit on 12/17/21, 11:48 AM

    Here's a simple one: You write out the requirements of the program in English and it spits out a program that implements it all.
  • by pcthrowaway on 12/17/21, 5:59 PM

    I hear that R is a good example of domain-specific programming that's largely used by non-programmers
  • by mesozoic on 12/17/21, 7:47 PM

    Probably something like you see in visual programming for kids training type things.
  • by etu on 12/17/21, 12:12 PM

    LaTeX

    I love the output it gives, but the language is quite horrible.

  • by karjudev on 12/17/21, 2:33 PM

    Definitely like a glorified Matlab
  • by aristofun on 12/19/21, 9:20 AM

    It would look like Ruby
  • by ModernMech on 12/17/21, 12:55 PM

    I’ve actually posed this question to non programmers. The language they come up with usually resembles natural language English. Alternatively it looks like a box/line drawing. It almost never looks like programming as we conceive it.
  • by Koshkin on 12/18/21, 12:02 AM

    Like CMake.
  • by thrower123 on 12/17/21, 12:59 PM

    QBASIC