However I will say, this source is unlikely to be useful to anyone beyond academic curiosity. So for the purposes of this project, the source license is “good enough”.
0. Freedom to run for any purpose? No.
1. Freedom to modify the program? Maybe; Microsoft can revoke it without cause at any time.
2. Freedom to distribute unmodified? No.
3. Freedom to distribute modified? No.
I'm using the free software definition [1] because it's simpler and clearer than the open source definition [2], which is usually considered mostly equivalent in practice, but as far as I can tell Microsoft's license satisfies zero of the latter's ten criteria, apart from perhaps non-discrimination and technology-neutrality. Most of the criteria of the open source definition have to do with distribution, which is not allowed by Microsoft's license.
[0] https://computerhistory.org/blogs/microsoft-research-license...
But to be fair/realistic, as an individual, I've never worried about their licenses (for myself, my friends, and family). The only time I did was when installing their software in professional/commercial contexts, that's where they do most of their enforcement as far as I know... and only when the scale is egregious.