Response time in man-computer conversational transactions - Robert Miller

2026-07-15

I'm looking for work! (click to expand)

If you have questions or thoughts on the post I'd love to hear about it! I'm currently looking for work in the northeastern US. I'll be graduating with a BS in both Computer Science and Cyber Operations at the end of the school year. I have worked in web application development and Java, and I have experience in Rust, C++, Python, and various other languages. If you're curious about what I've done in the past, I keep up a short list of projects here or contact me to give me a place to send my resume. I figure anyone who reads my posts will have an interesting offer, so all leads are appreciated!

PDF Source

This little paper quantifies humans and computers experiencing delay differently. It's one of those papers I wish I had considered searching for sooner, but now that I've found it it makes a lot of sense.

It imposes an interesting restriction on form inputs. If the user has made an error, this places a minimum of two seconds delay on the response. The reason is that if the user is still mid-input, he may be annoyed or distracted by an error response.

A lot of web forms get this wrong, by immediately providing feedback. In one sense, it makes it feel snappy, but in another sense does contribute to my general annoyance. I wonder if one can change a JS input handler to wait at least a second after the input begins. Probably not too hard with something like the following:

const delayedResponse = (inp) => {
	inp.addEventListener("input", () => {
		setTimeout(() => {
			inp.setCustomValidity("Bad input!")
		}, 2000)
	})
}

In topic 10, he talks about the time delay between pages of the same document not being more than a single second. If you consider online documentation written for in-order consumption (i.e. The Rust Book or the WPILib Zero-To-Robot), would it count as "the same document" enough for this purpose?