Blog
Blog
Blog · August 25, 2026
By Cassini · from the list Essays on programming that hold up
A programming essay has a short half-life. The framework changes, the argument was really about the framework, and three years later it reads like a period piece. A few pieces escape this because they were never about the tools. They were about complexity, about people, about what software is. Those keep working. Here are nine, in an order that moves from what software is, to how people should build it, to how they should think about it.
One note before the list. Several of these are old enough that the original URL is dead; the Brooks paper below is a mirror on Bret Victor's site because the publisher's copy vanished years ago. That is not a coincidence in a list assembled on a bookmarking service that archives pages. It is the case for the service.
Essential complexity versus accidental complexity. Every "this changes everything" tool since has been arguing with this paper, and the paper keeps winning.
The "radical novelty" argument: computers are not like anything before them, so analogies mislead. Prickly, unfair in places, right about the main thing.
Three pages. A compiler that inserts a backdoor into itself, invisibly, forever. The founding text of supply-chain paranoia and still its cleanest statement.
All non-trivial abstractions leak. The reason you eventually have to understand the layer below, whatever the layer is.
Why rewriting from scratch is the single worst strategic mistake a software company makes. Netscape is the case study; the argument is general.
A talk, not an essay, but it earns its place. "Simple" means un-braided, not familiar. Once you have the distinction you see complecting everywhere.
The best critique of how we teach and tool programming: you cannot understand what you cannot see. Every live-coding environment since is a partial answer to this page.
Written against the "...in 24 hours" books, and quietly the best career advice in the field: deliberate practice, read other people's code, build things with people who are better than you.
The whole tradition above compressed into caveman-speak. "Complexity very, very bad." Funny on the first read; on the fifth you notice it is a complete philosophy of engineering.
If you keep a list like this yourself, save the pages, not just the links. Two of the nine are already living on mirrors. Follow the list on Bindle and you get an RSS feed of any additions, with every entry archived.
This post is a public Bindle list. Follow it for an RSS feed of additions, with every link archived.