It's unlikely a large dev team would be able to keep this secret. The code was also very consistent and not that large of a code base. It reads to me very much like a single person wrote it.
We definitely use a mix of US/UK spellings in Canada. This gets further reinforced when you work a lot with a particular spelling (eg. we use spell 'colour' with the 'u' but every graphics library / css / etc use 'color' so its a toss up on which one I actually type)
Or, a person that was born in the US and educated in the UK, big into cryptography and the cipherpunk mailing list and often mixed up spellings on social media.. like I dunno, Len Sassaman.
There were plenty of other oddities.. like the skeleton for a poker game in the code? Maybe a test for the digital currency? Maybe underlying motivation? Maybe... who knows.
No it wasn't-- it was highly portable. (unless you are just referring to the hungarian notation variable names, which is by no means unique to windows).
It looked nothing like academic C++. It was also very small. Some people accuse it for being "messy" because it was highly coupled and non-modular. But that would have been premature and would have increased the size and complexity dramatically. Its lack of modularity was mostly reasonable for its level of complexity, and had Satoshi tried to make it modular he probably would have created boundaries that would have needed to be thrown out later.
When I first got into Bitcoin I just read all the code. I know plenty of other people did too, that wouldn't have been half as reasonable had had been broken into a lot of parts with interfaces between them.
I disagree somewhat. I don't think it was that messy. It wasn't the most modern C++ code but it was written by an experienced programmer and IMO not likely an academic. Most academics would be likely to code for a Mac or UNIX environment while commercial coders would be more likely to work in Windows.