You can get it by signing up here: https://training.kalzumeus.com/
This is a novelty approach, and it has no longevity. It doesn't last, and those aren't the users you want anyway. The webapps I use the most are the ones that actually offer something valuable to me. This sort of value is timeless. Google Docs is the first thing that comes to mind. I open Docs all the time on my own accord. Google don't whore themselves out with shiny new features because they don't have to. They wrote good software. It's a useful product.
I could see this strategy being more valid with web games, but as long as your webapp is a utility there's no dopamine burst you need provide users. Just help them get something accomplished and they'll be back.
Doesn't that undermine your argument and example?
I didn't mean it was "new functionality" every week like spam, but over time. That's the beauty of Labs for gmail.
What apps do you see that don't introduce new functionality to users over time that is successful?
I suppose during onboarding this is actually more important, at least until the point where the user decides "okay, I like this." I may have misunderstood the point. I just hate feature-focused development.
Perhaps you're reading more into it than the article actually said?
Another perspective would be that shoving every last feature in your new user's face on day one might be overwhelming and thus counter-productive. If you have a lot of functionality to offer, it might be better to start with the basics, and then adaptively introduce more functionality -- starting with the most relevant -- until the user is fully up to speed.