If you have several decades of experience, you should know that roadmaps are not necessarily about setting dates, they are about documenting the process.
I don´t get the reference to research, seriously. I have been working in research for years: 1. literally EVERYONE uses roadmaps, otherwise there wouldn´t be any way to communicate with supervisors, financiers and administration and 2. if we were close (i.e. ~24 months) to a stable release, which is a very common rhetoric here on the forum, you cannot refer to the process as “research”. Also I believe that you underestimate the process of constructing airports & buildings. It isn´t as straightforward as people believe and yes, people still use roadmaps.
I personally can´t see how you are not apologising the status quo. The example of Microsoft Word you use is rather the consequence of not having a proper roadmap and reminds me a lot of how it has been handled here: instead of saying “we need at least two years to develop a stable version” the process has been document in weekly updates and repeated hints that something will happen in the next few months. THIS combined with the expectations here on the forum is what drives pressure. Just have a look at this thread from 2014:
What I saw was that whenever a roadmap discussion came up the developers referred to the stress they are in and putting on themselves. That´s never a good sign, because it will lead to premature decisions. Quite the opposite, the discussion of a roadmap, handled properly, shouldn´t put pressure on the developers but provide them with the space they need to finalize the project. Maybe the whole team should take one month in a retreat without looking onto the code, Sometimes that is better than pushing hard every single day (#burnout).
And if it´s seriously a research project the roadmap should include several years before we start talking about releases. But that doesn´t mean there can´t be a roadmap.