I don't think all examination of qualifications constitutes a genetic fallacy. If anything it's closer to an argument from authority plus a closed-world assumption (i.e. I only accept arguments from authorities). But expertise and authority has a significant role in science, and not entirely an improper one imo. If I say that I prefer to hear expert analysis of nuclear reactors from someone with a PhD in nuclear physics, rather than from someone with a PhD in another field, that seems like a decent epistemological heuristic. The qualifications don't prove anything about the argument, but in general I expect people with relevant qualifications to know more of the relevant facts, so am more willing to defer to their judgment.
It's doubly the case when the original article author promoted the article as being by an "MIT PhD". If someone is explicitly invoking their credentials to add weight to an argument, investigating whether the credentials are relevant seems reasonable.
I disagree. The logic of the post clearly states that the source of the information is a reason to reject it rather than a problem with the information itself. This is a common problem with a vast majority of the negative responses to the "why I am not worried" article.
Remember, the poster says that the origin of the facts makes those facts bullshit. Textbook genetic fallacy.
What I'm discussing is evidential weighting and critical source analysis, not logical inference (I do happen to research logic in my day job, so I'm familiar with what constitutes valid deduction).
Since this isn't a scientific paper that presents data sufficient to support its conclusions, we can't judge it purely objectively. It presents itself as an engineering safety analysis. The first step in judging an engineering safety analysis is usually: was it performed by someone qualified to perform such an analysis? We typically want them to be performed by people who are both domain experts (in this case, nuclear engineers), and specifically people who are experts in assessing the risks in that domain, as well as in the general science of risk analysis. If a safety analysis is done by someone who isn't such an expert, it's quite rational to give it lower weight, because we aren't confident that they're familiar with all the relevant science and possible risks.
I mean, would you argue it's also a genetic fallacy that I take articles in Nature more seriously than I take articles in the Daily Mail? That I believe what Richard Feyman's lectures have to say about physics more than I believe what some random person on Geocities has to say? It just seems like good sense to me; considering the source of a claim is a good first step when deciding how much weight to give it. Bayesians would agree! It's possible that the Daily Mail will publish an insightful new analysis of global warming, but it's not very likely, and I would probably want to hear confirmation from a more reliable source before I believed it. Same here; this analysis is interesting, but I'd have more confidence in an analysis performed by someone who's actually an expert in the field, so until I hear one of them say that it's correct, I treat it with skepticism.
I'm arguing about the original post that I responded to, which makes the argument "this is not from an engineer, but a lobbyist and is therefore bullshit". That is a genetic fallacy.
If he had made the entirely reasonable point that you did-- that this might not be as trustworthy a source compared to another, or engaged the actual evidence used, I wouldn't have taken the time to critique it.
In other words, my disagreement with you has nothing to do with the weighing of credible sources, but the fact that the original post didn't do that.
Note that I haven't even brought up the fact that the post he actually responded to was a revised version of the article by nuclear engineers, thus invalidating much of his claims about it anyway.
It's doubly the case when the original article author promoted the article as being by an "MIT PhD". If someone is explicitly invoking their credentials to add weight to an argument, investigating whether the credentials are relevant seems reasonable.