After re-reading the "Human Wave" document, I've decided to devote a series of posts under the Vajra banner (since that's what it's most relevant to) to examining each of the suggested precepts within. Here's the first.
1. Your writing should be entertaining.
This seems like a given, doesn't it? But the problem I always run into with the dictum "Be entertaining!" is that there are at least as many definitions of "entertainment" as there are incarnations of it. There are people who find William Gaddis's JR entertaining, because the book tickles — entertains — a part of them that other books do not reach. Then there are people who see this 700-page doorstopper and just turn around and walk out of the room, because it offers them nothing they can find pleasure in.
If it's all subjective, then how is anyone supposed to know what will be entertaining? The answer, I think, lies in two things: the cultivation of an audience, and the good faith to follow through on your promises.
The first part is the harder of the two. No book exists in a vacuum; it's a contextual creation. It is a product of its moment in time as much as it is the product of its author (and since the author is also a product of that moment in time ... anyway, you get the idea). In the same way, audiences are a product of their moment in time. We have tons of people right now who read Twilight and The Hunger Games. That audience might not have existed in 1988, for instance; it might well not exist in 2018. On the other hand, a show like Max Headroom was way the hell ahead of its time in 1987 — and 25 years later, still is, which is really saying something. And it was as broadly entertaining as TV gets.
The problem is that it's often not possible to know what audience exists for a given book until it's written and circulated a bit. Movie producers are all too familiar with this: they really don't know what the next big hit is going to be, so they hedge their bets and make the next big hit as much as possible like the previous big hit. Some of this is unavoidable, but the formula's contracted to the point where every other film has a number after the title. (Note that this isn't exactly new — Gene Autry was a franchise for Hollywood as much as he was a talent, for instance - but it's reached such levels of waste and flummery as classic Hollywood would never have been able to guess at.)
What you must do, though, is meet the audience you have, half-way. If you write a book which makes little to no attempt to reach readers that exist in your moment in time, you will be shouting into a void. It makes no sense to turn your back on all that surrounds you in an attempt to create something new, new, new, because then you'll end up with something that speaks to nobody at all. It's cold comfort to tell yourself that someone, somewhere, in some unknown future, may possibly appreciate what you did ... but what fun is that?
This isn't the same as being of limited appeal. I know a number of authors who make no conscious concessions to reach an audience, and seem to do all right. They're not bestsellers, but they've carved out a niche for themselves, and they are able to address that niche of readers reliably. But they listened. They sought an audience, spoke to it, and used that as a way to further discover what they had to say without compromising their vision. There's a difference between the author being a creator and the author being a god. For a creator, there is no shame in telling a story worth sticking around for, for meeting his audience half-way — in short, for being entertaining.
The second point — following through on your promises — is something I see willfully violated in an age that confuses authorial irony for authorial contempt. If a book is marketed as a "thriller", and even written at least provisionally as one, it should at least deliver the goods. If it can't even do that — or, worse, if it refrains from doing that out of some foolish sense that to entertain the audience can't possibly be good for them — it will fall down. Rare indeed is the Sam Delaney who can give us a Dhalgren, or the Brian Aldiss who can give us Barefoot in the Head, two bits of experimental SF where the experiments actually succeed — but even then, that isn't to say those particular experiments are worth repeating, only learning from and applying to new work (even new non-experimental work).
The same goes for SF: how many jerk-out-the-rug endings, how many "I'm not really writing an SF novel" disclaimers, how many "experiments" that are really just thinly disguised expressions of annoyance or disdain or contempt for the audience? A good story, enthrallingly told and aglow with insight, is far harder to put together than most people admit, and I wish more people would follow through on that one simple promise when they are quite clearly capable of it.
So, yes: be entertaining, because nowhere is it written that being entertaining and being other things — intelligent, insightful, emotionally resonant — are at odds. If anything, being entertaining is one of the best way to be all of those things.