Dijkstra’s 1982 letter critiques APL by arguing the language shapes its users negatively and cannot be taught without a terminal.
Key Takeaways
Dijkstra wrote to Dr. Caplin in 1982 arguing APL’s cult status stems from how the tool shapes its users, not ease of use.
His core objection: APL cannot be adequately taught or discussed with pencil and paper, unlike languages grounded in formal notation.
Roger Hui counters that Iverson created APL notation as a communication tool before any computer implementation existed (pre-1963).
Hui presents two Dyalog APL derivations – Ackermann’s function and inverted table index-of – as proofs that APL supports formal, pencil-style reasoning.
The inverted table index-of derivation shows a practical performance gain: integer indices replace raw data for cross-table lookup, reducing overhead.
Hacker News Comment Review
Commenters are split: one argues Hui’s own APL examples actually validate Dijkstra’s point rather than refute it, since the notation remains opaque without execution context.
One commenter maps Dijkstra’s “gadget thrall” argument directly onto modern LLMs, noting users tolerate poor interfaces when sufficiently captivated by a tool.
Notable Comments
@Hendrikto: argues the post’s own APL derivations “validate Dijkstra’s points, instead of disproving them.”
@Almondsetat: suggests Dijkstra’s pen-and-paper standard was itself a historical artifact of pre-computer text culture, not a universal criterion.
