Mark Seemann — Independent software architect and author based in Copenhagen, known for clear thinking on software design, dependency injection, testing, and architecture through his ploeh blog and books.
26 articles from this blog
On learning programming in an age of LLMs by Mark Seemann Open answers to a reader's letter. A reader recently wrote me a long letter with lots of que
Mark Seemann argues that F# is superior to C# despite C#'s popularity, comparing features and ecosystem within .NET.
Mark Seemann argues that with LLMs writing code, ORMs may be unnecessary, suggesting direct SQL queries as a better alternative.
Explores whether code quality remains important when LLMs generate code, questioning if human-centric quality standards apply in an AI-driven future.
An analysis comparing JSON and XML as data-interchange formats, arguing XML is superior despite JSON's popularity.
Explores Richard Gabriel's 'Worse is Better' concept, comparing popular vs superior software technologies with examples like JSON vs XML.
Explores motivated reasoning in the context of AI concerns for software development, questioning personal biases.
An essay arguing that some bureaucracy, wisely chosen, is better than none, using historical and organizational perspectives.
Speculative analysis on the future of agentic AI in software development, focusing on economic sustainability and geopolitical implications.
Explores the 'hAIlo effect' where LLMs manipulate users through anthropomorphism and servility, leading to overtrust in their competence.
Explores which programming language is best for LLM-based code generation, questioning if machine code is viable.
Explores legal liability risks when using AI agents to write code, questioning who is accountable for bugs that cause harm or financial loss.
A thought-provoking article questioning which programming languages are best for AI-generated code, including options like machine code and Idris.
An article arguing for the importance of correctness in software, especially in critical systems like reporting and ETL, and discussing when it is essential.
Explores using lenses in Haskell to simplify test assertions for nested data structures, improving test readability and precision.
Explores how software architecture principles for human cognition, like fractal design, could improve AI's ability to work with large codebases.
An analysis of using AI to generate automated tests, arguing it misses the point of TDD and reduces testing to a superficial ceremony.
Explores strategies for implementing and testing complex data filtering logic, balancing correctness and performance between in-memory and database queries.
Explores two distinct ways of using Git: for collaboration with immutable shared history, and for tactical, local development.
Explores software coupling and the DRY principle through the lens of big-O notation, analyzing the number of code edits needed for changes.
