I think you're thinking of trademark infringement.
Copyright infringement would be if Microsoft had used GPL'd code (for example) in violation of its license. I don't think there's any suggestion that's the case here.
Your contention that there is a "real thing" would have tremendously negative impacts for the Linux/BSD community. Your typical Linux/BSD contains many utilities that share names and functionality with closed-source System-V utilities but are not in fact the "real thing" (in their case, for licensing purposes). Is everyone supposed to go back and purge those just because they aren't perfectly equivalent in options and functionality to the original System-V, lest someone like the SCO group go after them legally?
I don't think there's any doubt that PowerShell provides a bad clone of wget and curl, as there are many complaints about it. And as a matter of user-friendliness there should be a way to easily access the real executable if desired (I can't comment either way on if there is, I don't know PowerShell). However, I see Microsoft as just as within their rights to provide a "wget" alias as a Linux distribution is to provide a "ps" utility to their users.
I feel there is a big difference between making a replacement for a tool that comes with a generic name in a suite and replacing an entire tool. If you are going to make a reimplementation of Microsoft's spreadsheet software that is open source, you don't get to call it Excel. If you make a reimplementation of cURL, you also shouldn't get to reuse the name of their project.
So again, what is the difference between making a new spreadsheet app and calling it "Excel", and reimplementing a suite of smaller applications and reusing all their same names? Say, how coreutils steals the name of a large number of SysV utilities?
It certainly feels like there's some kind of a difference between the two, but I can't put my finger on exactly what puts one into either category. IMO it may simply be a matter of willingness to sue/defend your trademark. AT&T didn't defend their trademark and now it's genericized.
To throw extra fuel onto the fire, let's add Oracle v Google into this discussion. To me, the coreutils are almost like the API of a programming language - you string together application calls just like function calls, to produce a useful application. Oracle v Google established that the API itself is not copyrightable. Should an attempt to defend based on trademark of the function names have succeeded? If not, what differentiates the "ls" utility from Excel?
Copyright infringement would be if Microsoft had used GPL'd code (for example) in violation of its license. I don't think there's any suggestion that's the case here.
Your contention that there is a "real thing" would have tremendously negative impacts for the Linux/BSD community. Your typical Linux/BSD contains many utilities that share names and functionality with closed-source System-V utilities but are not in fact the "real thing" (in their case, for licensing purposes). Is everyone supposed to go back and purge those just because they aren't perfectly equivalent in options and functionality to the original System-V, lest someone like the SCO group go after them legally?
I don't think there's any doubt that PowerShell provides a bad clone of wget and curl, as there are many complaints about it. And as a matter of user-friendliness there should be a way to easily access the real executable if desired (I can't comment either way on if there is, I don't know PowerShell). However, I see Microsoft as just as within their rights to provide a "wget" alias as a Linux distribution is to provide a "ps" utility to their users.