Every time I see another startup trying use LLMs for code generation I sigh in despair. As AI technology improves and becomes better at producing code, what looks like a win in the short term will end up creating more and more code that has been created without a human going through the necessary thought processes and problem solving steps to build the theory of the software as described in this paper.
It's also why its critically important for companies to do what they can to retain the people who built the software in the first place, or at least ensure there's enough continuity as new people join the team so they can build their mental model by working alongside the original developers.
We might not be there yet (well we definitely are not) but it does not seem out of the question that within a generous 10 years we will have systems which can leverage graphs, descriptive language, interpreters, and so on to plan out and document and iterate and refine the structure of a problem and its architectural solution in tandem with developing the solution itself iteratively at a very effective level, given a sufficient explanation of the goals/problem - or more importantly/phrased another way, following the initial theory of a problem formulated by the human; the kind of documentation produced by such systems can also be more easily ingested by other non-human systems, potentially remedying some of the challenges with outlining/documenting/transferring the theory of the problem that humans have.
And what prevents a human from doing code review on such a system’s outputs? Now maybe your point was that the simple expense of a human’s time is the barrier, especially given that you were talking about the context of companies using LLMs to speed up their code production (read: eliminate cost centers), but in that case the errors that may come from poorly designed procedurally generated codebases just reads like bad project management to me for which the chickens will ultimately come home to roost; the companies which can successfully integrate such procedurally codegen engines while still maintaining strong design principles, maintainability, simplicity, etc ought to outcompete their competitors’ slop in the long run, right?
Having said all that, I think the more important loss is that the human fails to build as much intuition for the problem space themself by not being on the ground in the weeds solving the problems with their own solutions, and this will struggle to develop their own effective theories of the problem (as indicated by the title of the article in the first place).
The trouble with No Code is that your first paragraph is already my job description: I plan out and document and refine the structure of a problem and its architectural solution while simultaneously developing the system itself. The "sufficient explanation of the goals/problem" is the code—anything less is totally insufficient. And once I have the code, it is both the fully-documented problem and the spec for the solution.
I won't pretend to know the final end state for these tools, but it's definitely not that engineers will write natural-language specs and the LLMs will translate them, because code (in varying degrees of high- and low-level languages) is the preferred language for solution specification for a reason. It's precise, unambiguous, and well understood by all engineers on a project. There is no need to be filled by swapping that out with natural language unless you're taking engineers out of the loop entirely.
But: “And what prevents a human from doing code review on such a system’s outputs?” One word: cost.
At least in my experience, at least right now, it is more effort to review and correct, as doing from scratch.
https://www.goodreads.com/book/show/4594604-computing
is out of print, as is _Concise survey of computer methods_ and rather pricey.
Oddly, _Knowing and the Mystique of Logic and Rules_ (which has an even lengthier title after a colon...) has four entries at Goodreads and is listed under "P. Naur" and is even pricier, quite expensive on Amazon:
https://www.amazon.com/Knowing-Mystique-Logic-Rules-Statemen...
even as an ebook.
It would be more influential if it was affordably in print....
I think software engineers will drift towards only understanding the domain and creating tasks and then reviewing code written by AI. But the reviews will be necessary and will matter, at least for a while.
Note what Naur means with Theory here. Quoting from the paper:
"What will be considered here is the suggestion that the programmers' knowledge properly should be regarded as a theory, in the sense of Ryle [Gilbert Ryle, The Concept of Mind, 1946]. Very briefly, a person who has or possesses a theory in this sense knows how to do certain things and in addition can support the actual doing with explanations, justifications, and answers to queries, about the activity of concern."
This is not "theory" in the sense we sometimes encounter in colloquial speech in the sense of (exclusively) "assumption", especially not with the connotation "unjustified assumption". It is also not a set of rules:
"The dependence of a theory on a grasp of certain kinds of similarity between situations and events of the real world gives the reason why the knowledge held by someone who has the theory could not, in principle, be expressed in terms of rules. In fact, the similarities in question are not, and cannot be, expressed in terms of criteria, no more than the similarities of many other kinds of objects, such as human faces, tunes, or tastes of wine, can thus be expressed."
Yet, it plays a central role in programming:
"For a program to retain its quality it is mandatory that each modification is firmly grounded in the theory of it. Indeed, the very notion of qualities such as simplicity and good structure can only be understood in terms of the theory of the program, since they characterize the actual program text in relation to such program texts that might have been written to achieve the same execution behaviour, but which exist only as possibilities in the programmer's understanding."
Like hiring that one unicorn dev to solve X hard problem isn't a great "theory building" exercise. It can build theories for that one person, but without feedback they're never tested, they're never adopted by the whole team
So you actually NEED juniors, 'stupid' questions, outside points of view, and ways of openly and scientifically evaluating theories instead of defaulting to the authority of supposed experts. You also need to retain seniors who have context and a good historical working definition of the problem.
But a lot of teams are focused on just the next problem and "shipping it". Rather than using "shipping" to help the team develop a better theory of the problem.
The value isn't what's shipped, its the working knowledge of the team.
It is good if developer has already it, he is more productive then. But when he explicitly puts effort into gaining knowledge, then he does not deliver during that time so maybe he should not be paid for it.
I can't imagine a relationship between a manager and a developer where knowledge is valued higher than delivery. It could work only if the manager also believes in this value. I think he could believe in it only if he is sure that this project will pay off in the long run. In the era of a fast-changing world, he is putting the value of delivery and satisfying stakeholders on a higher rung.
Programming as Theory Building (1985) - https://news.ycombinator.com/item?id=38907366 - Jan 2024 (12 comments)
Programming as Theory Building (1985) [pdf] - https://news.ycombinator.com/item?id=37263121 - Aug 2023 (36 comments)
Programming as Theory Building (1985) [pdf] - https://news.ycombinator.com/item?id=33659795 - Nov 2022 (1 comment)
Naur on Programming as Theory Building (1985) [pdf] - https://news.ycombinator.com/item?id=31500174 - May 2022 (4 comments)
Naur on Programming as Theory Building (1985) [pdf] - https://news.ycombinator.com/item?id=30861573 - March 2022 (3 comments)
Programming as Theory Building (1985) - https://news.ycombinator.com/item?id=23375193 - June 2020 (35 comments)
Programming as Theory Building (1985) [pdf] - https://news.ycombinator.com/item?id=20736145 - Aug 2019 (11 comments)
Peter Naur – Programming as Theory Building (1985) [pdf] - https://news.ycombinator.com/item?id=10833278 - Jan 2016 (15 comments)
Naur’s “Programming as Theory Building” (2011) - https://news.ycombinator.com/item?id=7491661 - March 2014 (14 comments)
Programming as Theory Building (by Naur of BNF) - https://news.ycombinator.com/item?id=121291 - Feb 2008 (2 comments)
This seems especially relevant in open source, or in blog posts / papers, where we rarely have teams which continuously transfer theories to newcomers. Focusing on documenting "how it works under the hood" and helping others re-implement your ideas also seems more useful to break silos between programming language communities.
For example a blog post that introduces some library in some programming language and only explains how to use its API to solve some concrete problems is of little use to programmers that use other programming languages, compared to a post which would explain how the library works on a level where other programmers could build a theory and re-implement it themselves in their language of choice.
I also feel like there's a connection between the "Theory Building View" and the people that encourage rewriting your software. For example in the following interview[0] Joe Armstrong explains that he often wrote a piece of code and the next day he threw it away and rewrote it from scratch. Perhaps this has to do with the fact that after your first iteration, you've a better theory and therefore in a better position to implement it in a better way?
I also believe there's some connection to program size here. In the early days of Erlang it was possible to do a total rewrite of the whole language in less than a week. New language features were added in one work session, if you couldn’t get the idea out of your brain and code it up in that time then you didn’t do it, Joe explained[1] (17:10).
In a later talk[2] he elaborated saying:
“We need to break systems down into small understandable components with message passing between them and with contracts describing whats going on between them so we can understand them, otherwise we just won’t be able to make software that works. I think the limit of human understandability is something like 128KB of code in any language. So we really need to box things down into small units of computation and formally verify them and the protocols in particular.”
Speaking of Forth, Chuck Moore also appears to be a rewriter. He said[3] something in similar:
“Instead of being rewritten, software has features added. And becomes more complex. So complex that no one dares change it, or improve it, for fear of unintended consequences. But adding to it seems relatively safe. We need dedicated programmers who commit their careers to single applications. Rewriting them over and over until they’re perfect.” (2009)
“I’ve spent more time with it that any other; have re-written it multiple times; and carried it to a satisfying level of maturity.”
[0]: https://vimeo.com/1344065#t=8m30s
[1]: https://dl.acm.org/action/downloadSupplement?doi=10.1145%2F1...
[2]: https://youtu.be/rQIE22e0cW8?t=3492
[3]: https://www.red-gate.com/simple-talk/opinion/geek-of-the-wee...
Arguably, this is the entire spirit of academia, which mildly serves as a counter example, or at least illustrates the challenges with what you are describing - even in something where the stated goal is reproducibility, you still have a replication crisis. Though to be fair, I think part of the problem there is that, like you said, people focus too much on “documenting the artifact” and not “documenting how to produce the artifact,” but this is often because the process is often “merely” technical and not theoretical (and thus not publishable) despite being where most of the hard work and problem solving and edge case resolution and so on happened.
Edit: oh, and I would also mentioned, that the kind of comment you’ve described which focuses on why some process exists in the form it does to better explain how it does what it does aligns closely with Osterhout’s notion of a good comment in A Philosophy of Software Design.
http://literateprogramming.com/
I couldn't easily count the number of re-writes for my current project, but it keeps getting better, and each new iteration has had an updated architecture allowing for new features. When I re-wrote it as a Literate Program (first a .dtx, now a "normal" .tex) things got much more expressive and easier to work with.
I recently started building mobile apps using Flutter after a decade of developing apps using imperative programming languages and I’m really in love with the declarative nature of flutter.
Similarly for web development, I always loved HTML and so HTMX has been a boon for me. I’m using Go for backend, but I’ve been thinking whether I should move on to a proper functional programming language like Elixir with Phoenix since I’m liking declarative programming very much?
"I've seen this scan, what seems like a better digitized version of the same text, and this more formal-looking document, but they all refer to the essay in retrospect"
[Programming as Theory Building - Peter Naur](https://gist.github.com/onlurking/fc5c81d18cfce9ff81bc968a7f...)
A computer program is a "model" of some thing. For example:
float m = 1e10f;
float a = 9.8f;
float F = m*a;
if(employee is still employed):
float paycheque = getSalary(employee);
else:
float paycheque = 0.00f;For your argument to work, you need to prove that the original article is closer to a 1985 Cosmo article than it is to something like Clayton Christensen's 1995 article on Disruptive Innovation, which remains relevant today (or disprove one of the premises in my comment).