A quote by Ciro's Teacher R.:
Sometimes, even if our end goals are too far from reality, the side effects of trying to reach them can have meaningful impact.
If the goals are not ambitious enough, you risk not even having useful side effects so show in the end!
By doing the prerequisites of the impossible goal you desire, maybe the next generation will be able to achieve it.
This is basically why Ciro Santilli has contributed to Stack Overflow, which has happened while was doing his overly ambitious projects and notice that all kinds of basic pre-requisites were not well explained anywhere.
This is especially effective when you use backward design, because then you will go "down the dependency graph of prerequisites" and smoothen out any particularly inefficient points that you come across.
Going into such productive procrastination is also known informally as yak shaving.
There are of course countless examples of such events:
  • youtu.be/qrDZhAxpKrQ?t=174 Blitzscaling 11: Patrick Collison on Hiring at Stripe and the Role of a Product-Focused CEO by Greylock (2015)
The danger of this approach is of course spending too much time on stuff that will not be done enough times to be worth it, as highlighted by several xkcds:
Figure 1. xkcd 974: The general problem. Source.
Figure 2. xkcd 1205: Is it worth the time. Source.
Figure 3. xkcd 1319: Automation. Source.