You raised a good question: Can humanity sustain an effort for several decades?
I think the answer is yes. We're talking about only 4 decades (over estimating to be conservative) or maybe somewhat less. We aren't talking about rocket science. It's just building factories. The only challenge is delivering the equipment to the Moon over time (and improving robotics, which is already happening). We can do this in less than 4 decades, but let's stick with 4 to be conservative.
That's not so long. We have been planning, designing, building, and now operating the International Space Station for about 18 years and it will be extended beyond 20 years. So we're talking about only double the time of Space Station.
Another example is the ITER/DEMO fusion program. We plan to continue that program for a total of 43 years. We have gone only 9 years so far, but there is political will in Europe and elsewhere to keep going even though the leaders know how long it will take, so that's hopeful.
The US interstate highway system took almost 4 decades to complete. That may be a good analogy; we kept building it with taxpayer money because it kept providing value as we went along, so the taxpayers kept wanting more. Space industry will also provide increasing value as we keep expanding the in-space supply chain, making everything we do in space more affordable and more effective. As long as we keep wanting to spend US $16 billion per year to have a space program (as we have for many decades already) then the leaders of that effort will be motivated to keep making it successful, and they will see increasing value coming from space industry as we go along. (One of the points of my paper is that we can do this for only about 1/3 of the existing national space budget, so that gives us hope that we can really pull it off.)
More important than the continuity of a single project (like space station) is the continuity of progress in an entire technology area. Space industry is a technology area, not a single project. The space program is a global phenomenon involving many countries for many decades, not just NASA. Likewise, nuclear fusion research goes back for decades, long before ITER, and yet we are still committed to making progress all these many decades. Maybe we should study how to break down space industry into a series of projects with well-defined end points, each adding to the growing infrastructure in space. Each project might take only 1 decade and deliver a capability that becomes self-sustaining by providing value to the national space agencies and other customers in space.
Another key to sustaining this effort for 4 decades may be to make it an international effort with joint ownership via an international body, so it has resilience to withstand the end of each sub-project and the changing membership of international partners.
More thought is needed to answer your question fully, but I am optimistic. How to start space industry is a puzzle worth solving. The first papers written on it were incomplete because they assumed we would launch an entire factory into space, which was not realistic because of economics and politics. My (and my coworkers') 2013 paper tried to contribute to solving this puzzle by pointing out that we can use a bootstrapping approach rather than launching the factory all at once. We thought that put one piece of the puzzle into its place. We knew it was not a compete solution and we often talked about "what comes before bootstrapping space industry" to create the climate where it would be funded. We did not know. Some helpful work in that period included the ISU Summer Study called "OASIS" describing a network or propellant depots and space mining, and work by the space mining companies along similar lines, and how this can set up infrastructure that is self-sustaining and makes it easier and more likely to start bootstrapping a full supply chain. Many of us felt even this was still not a complete solution to the puzzle since it still requires large initial funding and investors may be unwilling to take on such risk. This new paper in the arxiv is an attempt to add another piece to this puzzle. I think it does get us closer to the complete solution. It argues that national space agencies can add space industry into a lunar outpost very adorably and they will get increasing science benefits as they go along. Furthermore, it argues that this has potential to solve global economic and environmental challenges so that adds further motivation for national space agencies to adopt this policy. It also says that I don't believe they will adopt the policy for some time, yet, and so an initial period of slower, incremental progress is necessary.
Rather than say space industry is impossible because of the remaining questions (and I know there are others in addition to the question you raised), we can get involved in answering them. This is an opportunity to think creatively. I believe we will answer them during this period of slower, incremental progress. Please get involved in helping solve this!
