It's also not clear that the story told about the 'primitve' folks and their weird mistaken beliefs is actually a historically accurate one. Which the wikipedia article also addresses in part.
I find a blog post that just regurgitates the 'cargo cult' metaphor as if the author had invented it, and also without doing any research into the accuracy of the actual story he's telling about "Papua New Guinea" -- to be kind of insulting to the reader, and lazy on his part.
As a small piece of proof that I didn't try to act as if I made it up--I linked to the Wikipedia post about the term cargo culting.
I have been thinking about cargo cults a fair bit recently in a (non IT) area.
EDIT: On further reading, I think I have a better idea now, but I'd still be interested in your thoughts.
"A comparison of status relationships in the different “fields” shows a definite common pattern. The dominant feature, which makes status relations among the Econ of unique interest to the serious student, is the way that status is tied to the manufacture of certain types of implements, called “modls.” The status of the adult male is determined by his skill at making the “modl” of his “field.” The facts (a) that the Econ are highly status-motivated, (b) that status is only to be achieved by making ”modls,” and (c) that most of these “modls” seem to be of little or no practical use, probably accounts for the backwardness and abject cultural poverty of the tribe. Both the tight linkage between status in the tribe and modl-making and the trend toward making modls more for ceremonial than for practical purposes appear, moreover, to be fairly recent developments, something which has led many observers to express pessimism for the viability of the Econ culture."
Kanban-style inventory replenishment is amazing for the automobile industry and manufacturing in general, but would be a disaster for package fulfillment for the likes of an Amazon warehouse. Logic Programming languages have been amazing at building Business Rule Expert Systems and Artificial Intelligence agents, but fail miserably at programming device drivers and OS kernels.
Likewise, the buzzwords that most people identify as meaningless buzzwords(eg. Object-Oriented of 15 years ago or Cloud Computing of 5 years ago or Big Data today) all have legitimate uses. And while most CEOs only pay lipservice to the concepts, you have no idea if those "buzzwords" are actually helping them out. You may think they look foolish, but they may be using it, in their context, very profitably.
The real lesson you should learn from Cargo Cults has nothing to do with avoiding buzzwords and the people that spout them. It is that you should learn how to identify appropriate vs inappropriate abstractions to the problems you face.
As a UX professional, I'm often told that we can just copy the menus in someone else's product, or the colors they've chosen, or any number of other decisions they've made about form and function.
It's why you see printers with touchscreens where buttons would actually work better, cars with vestigial body effects that do nothing, skeuomorphic elements where none belong, or "flat UI" where a bit of skeuomorphism would have helped the user.
There's no substitute for starting from user and business needs, instead of "what do the other guys do?"
I'd love it if more people started dismissing the idea of "cargo cult design."
These days I write "enterprisey" software and I've found that a clear simple design with controls that people understand is much better than an unconventional design (even if the unconventional design is more rational).
I suspect this is related to the classic principle of least surprise/don't make me think mantra.
That said I'm a back end developer who can handle some front end stuff and most definetly not a UX expert so my opinion is worth what you paid for it.
I've been bootstrapping my business, and when it comes to things like designing my homepage I can try to reinvent it from the ground up, or I can follow the cargo cult and copy the Stereotypical Startup Home Page (as patio11 calls it) and move on.
Unoriginal, yes. But it frees me from making that decision, and I know there is some sensibility behind it, even if it is a cargo-cultish thing.
I'm sure everybody does that, they just rationalize what they're doing as not just blindly copying, by coming up with "everybody knows" reasons for design features without any data to back it up.
The problem is that except where you are familiar with all the relevant technical details (and I mean technical broadly here), you end up having to fall back on other methods of evaluation. In the best case, you'll have the advice of an expert who is truly familiar with the relevant details, or enough of them to form good approximate insights.
Probably at least some of the time, though, you don't have access to that expert (or, worse, you're not equipped to distinguish an expert from a bozo). So, you back on heuristics... like social proof and examples of success...
The Wikipedia article on cargo cults is surprisingly fascinating: