[some time ago now, a comment from Arborman effectively asked what's this approach to housing you're talking about? I hope this is something in the vicinity of a useful response.]
At present, the COVID-19 recession has many people faced with eviction.
Well, generally, because the public sphere has not acted to house them. It hasn't done that because there's a construction of axiomatic justice which says, you must pay.
That in turn comes in with the triumph of wealth-concentration as the mechanism of organising society during the Carbon Binge; there were, before Great Harry suppressed the monasteries, alternate support mechanisms beyond your utility to existing wealth. There were, before Puritans demonized hospitality and generosity, constructions of virtue around charity as an act. (as distinct from the present "and if they weren't bad, we wouldn't need to do this".) There were, before the enclosures of the commons, villages as collective organizations to supply basic needs. Today, there aren't; it's nigh-impossible to have any form of collective organisation that isn't a for-profit corporation.
(What we're seeing with the US evangelical movement is a sort of mammonite memetic colonisation of the practice of religion. It's not a collective support mechanism in material terms, it's a ranched population of marks.)
If we're going to consider housing in a context of not being mammonites, we need to start with that axiom of justice; you must pay.
The simple reason for that axiom is that justice is presently constructed to guarantee wealth. This is a conscious and deliberate process and explains much of how wealth has had a presumption of virtue attached to it. (There aren't many working rationalisations for the inherent probity of keep the loot. Much better to avoid that question somehow.)
The structural reason for that axiom of payment is that once you introduce payment, you get relative advantage. If you iterate, you wind up with a small number of winners and a very large number of losers; dumb luck will do that. If you introduce relative advantage (often the dumb luck to be born with something that confers relative advantage), you get the same small number of different winners; different because you've introduced this bias. Because the iterations are generally slow enough to notice, you get widespread and then consensus rationalisations for why the bias is good; people who have it are sure it is, and they have most of the agency. (Plus people who don't have it still want it, so you get rationalisations about why it would be OK if they had it.)
Which is also how you get homeless people.
Which is a euphemism; "dying of exposure" is less euphemistic. "Lost some fingers to frostbite, struggled, then died, because that facilitates the relative advantage of someone with more than they can ever possibly use" is starting to get toward factual.
It's a brutal, brutal system. It's a brutal system even when people manage to remain housed at the cost of two-thirds of their income from working multiple unsafe jobs for low hourly pay.
So what would it look like if it wasn't brutal? If the axiom was something like people have value, rather than keep the loot?
People need, not just a place to sleep without fear, but a place to securely keep their stuff; people need a privy, and a place to bathe, and a place they can store and prepare food. It has to function all year round, in any weather. It needs to be a space you can clean without heroic effort; it needs to be something you can alter and decorate. (If you can't change it, it isn't yours, and if it's not yours you're definitionally insecure.)
So there's a whole lot of problems; how do you build that, in the century of angry weather? what do you make it out of, and how, so it isn't adding carbon load to the atmosphere? what heats it, what cools it, how do you ensure the heating the cooling and water and the sewage work even during external service interruptions? How do you make it quiet, and private, and have natural light, and not use up unreasonable ground area in what will need to be an increasingly dense urban landscape? How do you ensure this thing you've built keeps working across generations, because you're not going to want to rebuild housing again for a long while; it's going to be expensive. That's where Universal Design comes in, and should come in.
That's the general scale of housing problem; as a system, what kind of housing does society build?
As an individual, you have to ask mostly how do you pay for it? People need to move, it takes time to accumulate value, and the value in mortgages doesn't go to the house purchaser, so this isn't a question that has a sensible present answer.
I think the "pay for it" problems collapse into three general areas.
One is risk; two is durability, and three is mobility.
You manage risk collectively. That shouldn't mean a single massive collective housing authority, but a large number of just-large-enough housing collectives with specialities and geographic particularity. So you join one when you start being an adult; maybe you inherit a membership, maybe membership is passed to offspring of existing membership, maybe there's a joining lottery, maybe you just apply. But in all cases, three things are true; you get a membership in some housing collective. All the memberships are of required to be exchangeable, so if you want to move your existing membership can be swapped for membership somewhere else (and the range of values for housing collectives is constrained to some fairly narrow range to permit this). You can't lose your membership; it's not a transferable asset in the sense of something you can sell or lose in bankruptcy or not be able to pay for. (Which means there's going to have to be some public backstop when people get ill, hurt, lose their jobs, etc.)
Durability means a bunch of things; one of the things it means is you can't use an individual-scale market mechanism for this because most individuals know nothing about houses. The current housing market is about materials for producing houses and house producers buying land, it has nothing to do with the people in the house. So it cannot deliver value in the sense of habitability days per currency unit; the folks building houses want to keep doing that, if they can get the working life of a house under two generations that would be perfect.
We're going to need a new building code, new materials (and thus a lot of training AND some serious civil-power coercion to shift trades), and we're going to need a long-term, educated, skilled buyer. Oh look, a housing collective again; something that expects to be immortal, is trying to minimize its long term spend, and has the ability to buy value. (Which most individual house buyers aren't equipped to recognize, remember; you can't purchase what you can't identify.)
Mobility means people need to move; if you own a house today, moving is a problem because home ownership is only a good investment if you don't move. This is not of general economic benefit; rearranging communities of practice is a net positive, and systemic barriers to the free movement of labour are a net negative. Only one of those systemic barriers is produced by the effort to maximize cash extraction from housing, so we definitely have it, and we're going to have to abolish it. Which might mean restructuring all the financial institutions sitting on mortgages as their primary capitalization.
But, anyway; housing does not belong in the domain of markets, nor of profit. The question for housing is, "how many can we house, how well, for this much?" and that's a public-sphere question. Which means some set of public institutions as necessary to accomplish the task.