>
All of these are common social fallacies, and associating them with “geeks” in particular seems like stereotypingThere's no such thing as "stereotyping" geeks.
Geeks are already a stereotype -- they refer to specific category of people with specific traits (not everyone having exactly the same, but in a "family resemblance" kind of way).
If you want, you can say that invoking geeks is already stereotyping. But I don't think it's valid to say this is "stereotyping" geeks, as if geeks themselves weren't a stereotype.
So, "Geeks" already implies certain traits, and a narrow set at that.
In other words "geeks" is not like saying "programmers" (which is a much more varied group with hardly much in common). Attributing something to programmers in general would not be very successful (though, even programmers, are statistically more alike in some aspects between them than between the general population: e.g. for one, all of them use computers).
Stereotypes serve a purpose: they are ways to refer to collective traits and behavior shared by groups of people. It's expected that not everybody called an X is an X (we might mislabel someone). But it's not very logical to never make collective observations about groups. We do them all the time, in polling, in business, in politics, in medicine, in social studies, and so on. People are not exactly unique snowflakes, they also tend to have certain traits, and can be categorized into groups by the traits they have.
(And again, nobody expects the categorization to be 100% perfect. It's enough that it's close enough to be usable).