Needing to build buildings the size of football stadiums isn't exactly a significant issue, given we already do that with… football stadiums.
The hard part is "in space", where we need to solve both ISRU and getting there cheaply[0] for this to be anything other than vanity projects.
[0] Starship, if it works as advertised, is a wagon train to the stars — you need it to get going, but you don't want to do the Oregon trail in a wagon train when you have an airline or an interstate highway at your disposal.
Making ferris wheels airtight is somewhat pointless, but would be in certain regards easier in freefall than on the ground as (a) it doesn't need to start off spinning and therefore doesn't need to be stable until it's finished, and (b) there's not going to be the same concerns about metal fatigue because the force isn't going to constantly change direction from the reference frame of any given element.
If that can be balanced against (c) we've barely scratched the surface of space manufacturing and don't know what to expect, remains to be seen.
The London Eye is about 2/3rds of the size of a stadium. In addition, a ferris wheel has nothing like the mass you need to support for a space station.
There is no question that building a rotating structure with that kind of mass at that kind of scale rotating at those kinds of speeds is going to create significant novel engineering problems.
Those problems become even bigger when you consider that this structure has to be built in space (we've never built anything of this size in space, let alone tried to spin it up) and you have to solve problems like thermal expansion/contraction that will only get harder with scale.
> The London Eye is about 2/3rds of the size of a stadium. In addition, a ferris wheel has nothing like the mass you need to support for a space station.
You're being too literal, "stadium" isn't an SI unit and 120 metres is bigger than the 112 meters suggested in the article up-thread, so the technicality is not an important point.
Also, the core point is that the tensile strength required is nothing special. Indeed this is why the O'Neill designs are the size they are (8x32 km) and you only need to jump to mass-produced carbon nanotubes for the much larger and more ridiculous McKendree cylinder (size: Russia).
> kinds of speeds is going to create significant novel engineering problems
No doubt, novel things usually do even on the ground, that's why civil engineering is the discipline that it is — but can you name any not already faced by, say, suspension bridges?
In particular: what is speed supposed to do in this case, given its in space, and chosen specifically for 1g acceleration like everything on Earth is anyway?
The point of my comment was that there are implications for the human body with the rotational artificial gravity. The video sums it up pretty nicely with a live experiment in 6min.
I saw the video when it was new, and knew about Coriolis force since my childhood when my father (incorrectly) told me the thing about the direction water rotates when draining.
Its importance goes down when the parent object gets bigger compared to the contained object, which is why it's not important to draining but is important to hurricanes.
This was already treated in the 1970s NASA studies of space-colony concepts. In the first one they underestimated the problem, iirc, and had to recommend larger rings in the second.
It's hard to conceive how a space station with just some 100m of radius can even be useful. The artificial gravity is very likely far from a bottleneck.
Well, gotta start somewhere, and that's probably a decent size for a hotel.
I'm not sure how useful these would really be, of course. Only way to tell is to wait and see, unless you're working for Bezos or Musk and they're paying you to actually do the thing.
(Heck, part of me thinks they could be so terrible in practice that they are one of the fillers in the Fermi paradox, but I really should tidy up those thoughts into a proper blog post rather than ramble in a comment…)
It can be almost arbitrarily long on the other axis.
Though without specific use-cases to ask questions about — and I have none — you're right by default: just because it's possible, doesn't make it good for anything specific.
and the youtube example: https://www.youtube.com/watch?v=bJ_seXo-Enc