[TUHS] Re: What would early alternatives to C have been?

On Mon, Mar 10, 2025 at 10:41 AM Larry McVoy <lm(a)mcvoy.com> wrote: I'm with Arnold on this one in terms of preference, but I think there's a meta point that's interesting to consider: style localizations facilitate understanding and knowledge transfer across large code bases. Google has a codebase measured in the BLOC (B as in billion), and very strict style guides; they're really more hard and fast rules as opposed to a soft thing that the term "guide" tends to imply. Personally, I hated Google's style formats; but after a month or two they largely disappeared for me and I stopped thinking about them. I still thought and continue to think they're ugly, but the network effects of being able to `cd` anywhere in that massive monorepo and be able to read a file without the cognitive overhead of internalizing that this programmer preferred to put the braces somewhere else relative to the last programmer was a huge win, far outweighing my aesthetic objections. This directly contributed to allowing the organization to scale to tens of thousands of programmers working in the same code base simultaneously. Supported by tools like automatic formatters, folks concentrated on the problem at hand and not the minutiae of formatting. Uniform style also facilitated things like large-scale, automated refactorings; another huge win in evolving code to newer standards, fixing old problems, and so forth. So even though I didn't like the specifics of the local standard, I respected what it did for us. So I don't think the specifics of whether you put parentheses around every boolean expression matter as much as whether you aggressively commit to consistency everywhere and support that with tooling and automation. - Dan C.