--- title: "I'm running a programming contest · eieio.games" url: https://daily.dev/posts/i-m-running-a-programming-contest-eieio-games-oitrygale source_url: https://eieio.games/blog/im-running-the-icfp-programming-contest type: article source: "eieio.games" published: 2026-07-24T13:40:29.903Z updated: 2026-07-25T15:21:15.955Z tags: ["ai-agents"] reading_time: 2 upvotes: 0 comments: 0 language: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # I'm running a programming contest · eieio.games **[eieio.games](https://daily.dev/sources/eieio-games)** · 2 min read · 0 upvotes · 0 comments ## Summary The 29th annual ICFP Programming Contest has launched, designed by Nolen Royalty and collaborators. A key design challenge was making the contest resistant to AI agents running autonomously. The solution is a custom 2D programming language called 'littleman,' where programs emphasize spatial reasoning — characters physically walk over instructions to execute them and exchange data via drawn pipes. The contest runs for 3 days, and a public textbook on the littleman language is available to everyone. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Community take How the wider developer community reacted, aggregated from 2 discussions and 8 comments across hackernews, lobsters (as of 2026-07-25). **TL;DR:** The community is generally enthusiastic about the ICFP contest and finds the 2D spatial language concept interesting, though there's genuine skepticism about whether it truly prevents LLM use, with one thoughtful commenter theorizing it might work due to LLMs' sequential output limitations. **Sentiment:** 55% positive · 35% mixed · 10% skeptical **The case for** - The 2D spatial nature of the language may genuinely resist LLM use because LLMs produce output sequentially and cannot naturally iterate on a grid-based program. - The contest has a long, beloved history that generates nostalgia and excitement among experienced participants. **The pushback** - Nothing technically prevents someone from feeding the textbook and problem descriptions to an LLM, relying only on participant honesty. - The AI-resistance claim is theoretical and unverified — the designer hasn't proven it holds in practice. **By community** - lobsters (mixed): Lobsters commenters are excited about the contest but actively debate whether the 2D language design actually prevents LLM use, with one detailed theoretical argument that it might and others remaining skeptical. **Hottest debate:** Whether the 2D spatial programming language genuinely resists LLM agents or merely relies on participant honesty to enforce that. **Open questions** - Has anyone actually tested whether a top-tier LLM can solve non-trivial problems in the littleman language? - Could an LLM build a transpiler from a conventional representation into the 2D language, effectively bypassing the spatial constraint? **Highlights** > (Note: I am completely unaffiliated with this project, I just read the description.) Clearly, one can point point a top-tier LLM at any problem and expect it to do a job comparable to a low to middling skill programmer. But I do think that this particular programming language is specifically designed to be resistant to LLM success. One of the fundamental limitations that LLMs have is that they produce their output sequentially. They spit out a series of characters (or tokens) one after the other. This language is designed to require two-dimensional thinking. I would find it difficult to believe that any human could sit down and write out a non-trivial "program" in this language working left to right, top to bottom. The effective way to work is to take a big grid and start filling it in, making errors and corrections as one goes along. But this is something that an LLM cannot do naturally. Sure, I could build a tool to consume some other representation and spit out executable code in this language (although, it would be somewhat tricky to do so). And I'm sure someone seriously dedicated to making an LLM write programs in this language, could get an LLM to build such a tool. But simply showing the language definition and problem statement to an LLM is not going to be particularly effective. Important note: this is me just theorizing based on general principles; I have not tried it. > — [mcherm on lobsters · 1 points, 1 comments](https://lobste.rs/s/yajc8q/i_m_running_icfp_programming_contest_2026#c_dqjjxd) > Honesty and/or a desire to have fun programming by oneself? > — [abhin4v on lobsters · 1 points, 1 comments](https://lobste.rs/s/yajc8q/i_m_running_icfp_programming_contest_2026#c_mdz8do) > Every year I forget that ICFP is coming up & fail to put a weekend aside to have fun programming. Clearly I should be putting it in the diary a year ahead... > — [pja on lobsters · 3 points](https://lobste.rs/s/yajc8q/i_m_running_icfp_programming_contest_2026#c_8fjuib) **Source threads** - [hackernews](https://news.ycombinator.com/item?id=49036592) · 1 points · 0 comments - [lobsters](https://lobste.rs/s/yajc8q/i_m_running_icfp_programming_contest_2026) · 9 points · 8 comments ## Similar posts on daily.dev - [Langjam Gamejam](https://daily.dev/posts/langjam-gamejam-3wonyfcyd) · Lobsters · 2 upvotes · 0 comments - [I'm Coding by Hand](https://daily.dev/posts/i-m-coding-by-hand-kdozpquzj) · Hacker News · 4 upvotes · 0 comments - [The 29th IOCCC](https://daily.dev/posts/the-29th-ioccc-zwoipulya) · Lobsters · 0 upvotes · 0 comments --- Tags: [#ai-agents](https://daily.dev/tags/ai-agents) [View this post on daily.dev](https://daily.dev/posts/i-m-running-a-programming-contest-eieio-games-oitrygale)