Tmax is claiming Tmax Window uses a microkernel. Windows NT uses a hybrid microkernel/monolithic kernel. Linux uses a monolithic kernel. Tmax says it will run both Windows and Unix apps. So, with this information, these are the possible truths to this situation.
1. Tmax is lying about the kernel architecture, and Tmax is Linux using Wine. Possible.
2. Tmax is honest about the kernel architecture, and Tmax is MINIX using Wine. Less possible.
3. Tmax is a ReactOS hack that is using a POSIX compatibility layer for Linux compatibility. (I can't remember how our POSIX is working right now, never really tested it.) Most likely.
4. Least likely, a small Korean comany none of us has ever heard of has pulled off in a few years what it's taken an international open-source project over a decade to do.
Too bad there's no photos of the command line. Then we'd be able to tell if it was running CMD.exe or Bash.
We can assume that they aren't running at 100% compatibility because neither ReactOS or Wine run at 100% compatibility. The only way to pull that off would to do dirty reverse-engineering, which is a big no-no. On the bright side, if Tmax did go dirty, the Wine guys would have someone else to be suspicious about instead of ReactOS.
They ran Starcraft? Big whoop. Starcraft is listed as a "Silver Application" on Wine's compatibility database. We can run it too.
So there's a public beta in November. I'm guessing if they are stealing ReactOS code, then they've hacked together a new GUI (or skinned over explorer_new) and wrote a few proprietary programs with no regard to Windows compatibility. Assuming Tmax Window takes as long to develop as a Windows release, then a year from then we'd see the Gold release. We're almost at 0.4, and by November 2010 we should be around 0.5 or so. It seems awfully convenient that Tmax Window will go gold just as we're hitting a beta release that is stable enough for everyday use.