The industrial military complex is built on funding for proxy wars with Russia. I wonder if the issue this time is that they are worried that with Russia directly involved instead of by proxy, this war may end up breaking Russia if they lose. Dismantling the perpetual antagonist that motivates further funding of the war machine is not in the interest of those who make money on wars.
While a broad concept, in the context of your question, science is a metod to derive knowledge from observations.
Alternatives to the scientific method is to guess or to obtain knowledge from others. (Most other ways I can come up with, e.g. “religion” can still be sorted under these two.)
Obtaining knowledge from others is great, but may not always be available, and the quality of the knowledge derived this way depends on the reliability of the source.
For the other alternative, every sensible metric shows how science is a better method than guessing to derive knowledge.
As a different, more techy, solution that can work depending on the people you collaborate with, is to use a hosted Git service for collaboration (if you want to stay completely open source, a self-hosted GitLab).
Then, change your publication workflow to write in Markdown, ReST, or one of the other ascii formats that previews correctly, and set up your CI to render the documents automatically into, e.g., pdf:s using a converter. There are all kinds of converters from Markdown/ReST -> docs, presentation, etc. formats that are as competent - if not more so - than the usual office suites. This setup offers both online editing in the GitLab instance and offline by local cloning of the Git repo.
The side effect is that this system very seriously records and preserve your document history. You can see exactly who, at what point, changed, added, and removed things. For some types of documents, this can be very important.
After checking that you can open port 53 udp yourself with, say, nc (which you tried), strace the binary that tries to open port 53 and fails, and find the system call that fails. You can compare it with an strace on nc to see how it differs.
If this doesn’t clue you in (e.g., you see two attempts to listen to the same port…) Next step would be to find in the source code where it fails (look for the error message printout) and start adding diagnostic printouts before the failing system call and compile and run your edited version.