I'm sorry but I'm still failing to see how worldbuilding = bad. I can understand one's preference for more or less worldbuilding (and thus picking a playstyle that speaks to that preference)... but so far, and maybe I've missed it in this enormous thread, what I haven't seen is a good argument for why it should be considered "bad" as a default. I mean it's a hobby, it's supposed to be fun and if for some people fleshing out their imaginary world or reading about the numerous details found in a published imaginary setting increases the enjoyment of participating in the hobby, I can't see how that could be considered a bad thing, especially when (in the case of a published world it can be ignored by those who don't want it) or in the case of a homebrew isn't impacting the game in a negative way for the players. Which brings me to another point...
How much worldbuilding should be considered excessive... well IMO, that's totally dependent on the players of the game. I know my players, I know what they're likely to seek information about, what they are likely to investigate in a given situation and even what they are likely to do in a general sense (and if I honestly have no clue I will ask them outside of playing time about their plans before our next gaming session)... so while the details I create may seem superfluous or excessive to someone not running for my group, they aren't for me because even though they may not be used in the game session they are providing a buffer that makes running a game a more comfortable experience for me with my particular group. I mean really that's the only measure that matters to me... whether the worldbuilding that has been undertaken (irregardless of actual direct usage in tonight's particular adventure) has enhanced the play experience in some way.
Well, I agree that you are right and that it will vary greatly from person to person and from gaming group to gaming group. And I don't think that worldbuilding is inherently bad. The idea from the article in the OP is that you do not need to detail an entire world before you even begin to write. Now, the article is more about fiction writing than RPGing, so right there, there's a bit of loss in relevance.
But to translate that idea to RPGing, I think it's a valid concern. How much do we need to detail ahead of time? Again, this will vary, as you mention. But in general, I think that most GMs can likely get away with less worldbuilding than they think is necessary.
I used to do a lot of worldbuilding ahead of play. I found it to be entertaining in and of itself. But I do think it tended to lock me in to what I wanted the game to be about. Not completely, but at least partially. It didn't matter if my players expressed little interest in Cool Idea A that I introduced....I would find a way to get them to engage with it. Now, very often the players wouldn't mind and they'd accept that's the way the game is going, so let's go. But what if I didn't feel the need to shift things back to what I wanted? What would have come up? Did I miss out on something spontaneous and more in line with what the players would have wanted, and which was at least as cool as Cool Idea A?
I mean, when people talk about what makes a good GM, adaptability is one of the top qualities that is mentioned. So I think that's what it boils down to; worldbuilding should be a tool that can be used to help the game. It should not be the point of the game.
These days, I still come up with ideas ahead of time. But I keep them loosely defined. I keep things flexible so that I'm not so married to my pre-written material that I can't let it go in favor of an idea that comes up spontaneously in play.
This applies to backstory, too, which I think also gets criticized in a similar way. So much of the backstory that is done by a GM won't actually come up in play. Sure, it may inform things that impact play, especially GM decisions, but it remains an unknown factor from the players' perspectives. So again, best to keep this stuff minimal. Have a basic idea of the lineage of the current king and how he rose to power. You don't need a full family tree and detailed history for this guy. I don't think having that info is inherently bad...but how else could that time have been spent? Perhaps there is a more productive way to prepare for the game than to write up this level of detail.
So I think that the criticism has some merit. I don't think it's anything like a universal truth. Nor do I think that it tends to be a huge problem in most cases because most players and GMs will likely talk about this stuff, and try to resolve any problems.
But I do think it's something that each GM should keep in mind. Something to be aware of when you are GMing and working on worldbuilding or writing backstory. It's a potential pitfall, and it can be avoided. But you have to know about it to avoid it.
Now, when it comes to published products and the amount of space they devote to worldbuilding...in that case, I think a variety of products is best. Something like Vornheim the city guide which is entirely utilitarian in its approach is just as valid as something like the Grand History of the Realms, which is purely setting background. I don't get the desire to limit such products to one extreme or the other.