I suspect my comment comes too late, but, having finished Professor Lynch's book and enjoyed it, I am motivated to write, even if no one reads what I say.
Professor Lynch does commit both the naturalistic fallacy and the is/ought fallacy. But he does not conflate them. Indeed the key to his argument is that he doesn't. Instead he applies a pollster's or psychologist's empirical approach to the "ought" problem. Maybe the better way to put it, he applies a kind of Delphi method to the "ought" problem. But he does not use a naive "intuitionist" or "sentimentalist" approach, as does Hume. Professor Lynch is more thoughtful than that. So Mr. Heersinki's criticism is solid, and would be even more so were he to make the following distinctions in Professor Lynch's methodology.
Rather than committing straight-forwardly the naturalistic fallacy, a simply confusion between "is" and "ought," Professor Lynch is more provocative. He repeatedly founds his "ought" statements on what he believes most people think or almost instinctively feel. So he asks if he or we would like to be brains in vats, and he is correct to say neither he nor I nor most of the rest of us (probably) would choose that option. He argues therefore, what most people think is an indication, empirically valuable data, of whether the truth is an objectively existing "ought." In other words, he is saying, "XXX million people must be on to something for them to all feel alike."
But it is here he makes a fundamental error, not of logic but of empiricism. Though one's first reaction is to say that it is a deep fallacy to implicitly assert the truth of a proposition based on what some number of people thinks it is true, I do not believe the truth value of widely shared feelings and beliefs can be so easily dismissed. Certainly, the earth is not flat, no matter whether everyone in the world thinks it is, and it is highly likely the Theory of Evolution is true, regardless of the large number of Creationists.
But I do not automatically dismiss the views of Creationists. Rather, a view held by so many motivates me to think, not dismiss. Similarly we also know empirically there is "wisdom in crowds," at least as to certain questions. When a great many people believe something to be true, it deserves our attention, though that alone is not sufficient to support agreement.
That attention it deserves arises from the demonstrable truth that what a great many people think often turns out to have substance or partial truth, though hardly all the time. Put another way, what a great many people believe or feel are legitimate data for addressing "ought" questions. If one is asking whether finding and acting according to the truth brings happiness or a certain satisfaction to many people, instrumentally or as an end in itself, an empirical finding that the truth does bring those things cannot be ignored as irrelevant to certain, important "ought" propositions..
If it is empirically supportable that finding and acting according to the truth brings happiness or a certain satisfaction to many people, then one to say to many (most)(all) people, "You ought to seek the truth, if you have as your goal happiness or a certain satisfaction." Our study of human behavior empirically supports this contingent "ought" assertion.
In that way, the is/ought distinction can be substantially blurred or even eliminated. Professor Quine does something similar, I believe, in his criticism of skepticism, Hume, and the analytical/synthetic distinction.
What remains then is for Professor Lynch is to show empirically how many people actually believe as he says they do. His argument based on what you and I might believe, explicitly or tacitly, just isn't remotely good enough to even suspect his "ought" conclusion is true. He can substantiate that conclusion only if he can show empirically that most do indeed find happiness or that certain satisfaction when finding and acting according to the truth of the world or the semantic position that asserts it.