(…although I doubt he consciously intended it that way!)

I'm pretty sure Eliezer's "Death With Dignity" post was an April Fool's joke.

Dawkin's earlier mistake (since recanted) re the Selfish Gene

What mistake is that? I know there's a couple details that have been updated (e.g. the replicator / vehicle of selection distinction, from The Extended Phenotype, improved things) but overall he, like many other evolutionary biologists, still thinks it holds up really well

Coyne had complicated feelings about Gould (but loved Gould's sometimes-collaborator Lewontin, who was his supervisor). Here's Coyne responding to Trivers' "Vignettes of famous evolutionary biologists, large and small":

...Gould comes in for the greatest thrashing, especially for his flawed analysis of Samuel Morton’s cranial-skull data described in The Mismeasure of Man. Apparently motivated by his anti-racist sentiments, Gould apparently didn’t look too closely at what Morton actually did before accusing him of unconsciously manipulating data. (Gould’s book, however, is well worth reading for the other stuff.)

Trivers also goes after Gould for his (and Eldredge’s) theory of punctuated equilibrium, and here I think he’s right. If you construe that theory as being not just about patterns in the fossil record but about evolutionary process—about traits being molded by species selection—then Gould was simply wrong about that, and Trivers’s conclusions are correct. Yet, in the last chapter of my book Speciation (coauthored with Allen Orr), I think we make a persuasive case that species selection has operated in nature, and has molded the frequency array of characters that we see around us (i.e., what proportion of birds, among all birds, show sexual dimorphism for color?) Gould’s mistake, I think, was to suggest that species selection could somehow create adaptations themselves rather than just affect the array of existing adaptations.

When I think about Gould’s scientific achievements, I come up with very little concrete discoveries he made that are of any note. But he was seriously important in restoring paleobiology to a respectable discipline, for he had the rhetorical and writing skills to revive that field. And that, at least, is an accomplishment worth celebrating. Further, Gould’s Natural History essays and other popular pieces were always interesting, if sometimes tendentious, and surely helped awaken the public to the marvels of evolution.

As for Gould as a person, I had little use for him. In my experience the man was arrogant, preening, and completely lacked empathy, especially for us poor students trying to ask him questions. He often treated people very shabbily. Gould was a smart man and an eloquent man, but not a nice man. But we’re used to such people in science.

Note that the "Williams revolution" that Gould is widely seen as having stolen (and which Tooby accuses him of having stolen) is not what Eliezer Yudkowsky says it was. It had nothing to do with the argument in Full House, nor to do with Williams' equilibrium-amount-of-information argument, which is not at all the most widely-remembered part of Adaptation and Natural Selection. (I actually think Gould's argument makes sense, or partial sense, insofar as it's a response to the lay public's impression of the big-scale evolutionary history of life on Earth as having been driven to climb a ladder from simplicity to complexity. Why did fish evolve after single-celled organisms?—Well, it could hardly have happened the other way around, could it?)

The revolution in question is about criticisms of pan-adaptationism ("panglossianism"; "ultra-Darwinism"), i.e. the habit of assuming every trait you can think of in an organism was naturally selected for some purpose [and the habit of speculating what that purpose was]. (BAHfest, the "festival of bad ad-hoc hypotheses", invented by SMBC, also mocks that habit, which is still around despite the best evolutionary biologists' best efforts; e.g. I was in a first-year bio class where the TA tried to illustrate selection thinking by speculating that caterpillar mouths open sideways so they can walk along the edges of leaves and eat; sure, maybe, but maybe it's relevant that insect mouths in general do that, and maybe vertical vs. horizontal makes little fitness difference anyway.)

Williams gave the good example of a flying fish returning to the water. Is it fitter to eventually fall back into to the water than to not? Probably, sure. Does that mean it's an adaptation? No, of course not, it's just the default you should expect because of gravity. Indeed the first sentence in Adaptation and Natural Selection was (1966):

Evolutionary adaptation is a special and onerous concept that should not be used unnecessarily, and an effect should not be called a function unless it is clearly produced by design and not by chance.

Gould and Lewontin also criticized pan-adaptationism in their paper, "The Spandrels of San Marco and the Panglossian Paradigm: A Critique of the Adaptationist Programme" (1979). They gave a memorable metaphor, and coined the term "spandrel" for traits which are not themselves adaptations, but developmental side-effects of other traits (e.g. male nipples, and female orgasm, in mammals). The metaphor came from architecture: the spandrel (or pendentive) is the tapered triangular bit that results, if you put a dome on top of rounded arches; the spandrels in St. Mark's Basilica have such beautiful mosaics that a naive tourist might assume the spandrels were put there for the mosaics; but they'd be wrong. I agree Gould and Lewontin should have cited Williams, but I'm glad they gave us the metaphor; it's almost as beautiful as the spandrels themselves. They also coined the usage of "just-so stories" (after Rudyard Kipling) as an insult-term for untestable post-hoc speculation.

The context of Tooby's letter to the editor is that Gould had been accusing evolutionary psychologists of being Panglossian just-so storytellers. Tooby was replying that they knew perfectly well not to be Panglossian; Williams was their doyen after all; that they were in the business of testing hypotheses' predictions, not just speculating about the past; etc.

Anyway, as for Gould himself—I wouldn't say never read him, but I would say read him with a big shot of salt; and yes, if he implies he's revolutionizing something, don't believe him. As Maynard Smith says, his essays are often excellent. But e.g. he way-overblew how big a deal punctuated equilibrium would be even if true (see the chapter about it in Richard Dawkins' The Blind Watchmaker). And so on.