I've been waiting for someone "in the know" to tell us the state of future
US ECLSS. I knew it was bad. I keep hoping that the US will find a way to
get ECLSS up earlier than the "HAB", but it doesn't look promising.

I'm not in the know, but I can provide mostly-uninformed speculation
;-).

The thing about ECLSS is that it requires various ducts, connections,
software, and the like. It isn't just like an experiment rack where
you plug it in, provide it with power, and it is ready to go. So I'm
not expecting to see a retrofit which involves running a water line
(for example) to an existing rack in the Lab or such. I could be
wrong on this: maybe the existing lines go more places than I realize
(I suppose Node 1, at least, would need to have plenty of them so that
they can go to the eventual Hab).

Could one design ECLSS in a more "appliance" like way? For example,
having the astronauts carry the water over in a bag and put it into an
oxygen generator manually from time to time? Possible, but it is a
lot easier to do that kind of task without spilling in 1 g than
weightlessness. Certainly the oxygen candles are an example of this
kind of solution.

Could one run semi-ad-hoc lines fairly easily using things like
flexible tubing (all the rage in terrestrial plumbing these days, at
least for some things)? Maybe, although I'm sure it wouldn't be a
small task to run it behind the racks and it would perhaps get in the
way if it is in front of the racks. I guess the biggest problem is
getting through a hatch. The standard ISS solution, as I understand
it, is a through-hull on each side, followed by an EVA to hook them
up. Well, if the US LAB (for example) doesn't have the through-hulls,
adding them on orbit seems .... unlikely. The Mir solution of course
was just to run things through the hatch and having to unplug it if
you ever need to close the hatch. I'm not sure even Mir did this for
things like water and gas lines. And of course in cases like the
crash and air leak, all the cables through the hatch were considered a
problem.

But yes, if anyone actually knows more about studies of
ECLSS-before-Hab, I'd be eager to hear more about what problems of
this sort came up, and what the proposed solutions were.