First, let me just say that I am not a lawyer, and all I can say on this topic is what I've found online, where non-lawyers tend to give out substantial quantities of bad advice, and there are few legal scholars volunteering free guidance. If anyone can post a more definitive answer to my questions, I will happily withdraw this and chose a better answer, once I've got a reasonable selection to chose from.
What follows is my interpretation of the information I have found. Hopefully, minus the noise.
Let's start with the "is Microsoft compliant with MIT license?" part of this question, since that is the title of this thread.
According to MIT License - Wikipedia, there is no one MIT license. They point out that MIT has released code under more than one permissive license, however, all of the variants listed, include the year of creation/publication, but we must keep in mind that these licenses date back to the 80's when US and other laws were substantially different, and the presence and form of the copyright notice were strictly proscribed.
The Open source Initiative only lists one MIT license which does include the year.
Microsoft's LICENSE.TXT file, for the Core repository, starts with the line "The MIT License (MIT)". It's not at all clear what variant or version they are referring to, but the parenthesized 'MIT' happens to be the SPDX short identifier for the MIT license, and since it is redundant, I assume they are referring to that SPDX, which takes us to the OSI. The fact that they do not comport with the MIT license published by the OSI, suggests what they really have is a variant of some MIT license.
Wearing my pedantic cap, I'd say the OSI and most of the various license authors have generally failed to adequately separate copyright concerns from license concerns, but I know from previous lurking on their various email lists, that many of the legal experts, claim that the copyright notice itself is integral to any OSS license, as you must first establish that; you are granting permissions based on your legal ownership of the work. Without it, there's no grantor for the grantee to be bound to contractually. It is the MIT license after all. On what basis would MIT be granting anyone a license to someone else's work?
So a statement of who is the grantor of the license, seems to me to be required for any contract to be valid. In this case, the grantor can only be the current copyright owner, hence the copyright notice. Keep in mind that the distributor of the source code in question, may not be connected/related to said owner, so including this information in the license, attached to the source, seems perfectly reasonable.
It is my opinion that Microsoft should make it clear, that what they have is a variant of the MIT license, not the MIT license itself. So the answer to my question is a qualified NO. Qualified in the sense that, I am just not a lawyer and there are no legal restrictions, I am aware of, on the use of the MIT license published by the OSI.
But Microsoft includes a copyright, it's just not of the form suggested by the OSI or used by similar licenses. So why the variance?
That brings us to "What are the legal consequences of each of the two forms of copyright notice?".
As pointed out in https://www.copyright.gov/circs/circ03.pdf (emphasis are mine):
Although notice is optional for unpublished works, foreign works, or works published on or after
March 1, 1989, using a copyright notice carries the following benefits:
• Notice makes potential users aware that copyright is claimed in the work.
• In the case of a published work, a notice may prevent a defendant in a copyright infringement
action from attempting to limit his or her liability for damages or injunctive relief based on an
innocent infringement defense.
Copyright Notice 4
• Notice identifies the copyright owner at the time the work was first published for parties seeking permission to use the work.
• Notice identifies the year of first publication, which may be used to determine the term of
copyright protection in the case of an anonymous work, a pseudonymous work, or a work
made for hire.
• Notice may prevent the work from becoming an orphan work by identifying the copyright
owner and specifying the term of the copyright.
So two of the listed benefits, in the US at least, and probably others, mention establishing the date of publication. I seem to recall that copyright begins at the point of creation, not the point of publication, but establishing the creation date is difficult without some form of public disclosure. It seems that Microsoft is willing to forgo the benefit of a date established in the copyright notice. Perhaps they feel their publication process is adequate to the task of establishing a year of publication? Perhaps they prefer the potentially later publication year, to be the point at which their legal term of copyright begins? I can't say, but those are the only reasons I can think of at this time.
I think the answer to #2 is yes, if you wish to reset the start of the remaining term of copyright. Now this may bring us to another of Microsoft's motivations (only speculating of course). If you modify a work in a given year, and you're tracking years in the copyright notice, then you must update the notice to include the years of modification, otherwise you run the risk of not gaining the benefit of the potential extended copyright term. Now there may be other ways to establish that the work was updated, and attribute the changes to the listed copyright owner, but why complicate that process? Either update the notice in every year the work was modified, or don't post the year in the copyright notice to begin with. For a large organization like Microsoft, ensuring that a copyright notice is updated with every change, would require the build system inspect every changed file for an accurate year and either fixing the notice or stalling the build for manual intervention. The former would require tooling and build time, while the later is not worth contemplating. It's far simpler to enforce a fixed static copyright notice in every source file.
For #3, I think the answer depends on your jurisdiction. In the US, it seems redundant to me, but there may be countries where the symbol takes precedence. I believe that the copyright symbol is common to many, if not all countries, but the word "copyright" might not be so obvious in some languages. I suppose a little redundancy doesn't hurt, but I can only speculate why Microsoft chose to use both.
#4 just restated the title, and I think I've made it clear that they are using a variant. It's still not clear to me why they chose to call it "The MIT License (MIT)". Perhaps someone will enlighten us, for history's sake?
In my search for information, I stumbled across this excellent page written by another non-lawyer who seems to have done some good research, where I was reminded that, in the US, you can only sue someone for infringement if you have registered your copyright with the US copyright office. I think we can assume that Microsoft does in fact register in most, if not all jurisdictions that require it. Perhaps it is the US registration they rely on for establishing the publication date?
Apologies if I have asked more questions than answered here.