From: Tim R. <ti...@pr...> - 2015-03-03 17:51:39
|
Pete Batard wrote: > From [PATCH 1/3]: > > > libusb/os/windows_nt_common.c | 572 > > libusb/os/windows_nt_common.h | 67 > (...) > > + void win_nt_destroy_clock() > (...) > > + int win_nt_handle_events(struct libusb_context *ctx, struct pollfd > *fds, POLL_NFDS_TYPE nfds, int num_ready) > > I'm gonna have a very *subjective* objection to suffixing anything with > "NT" in 2015! > > The use of NT was already obsolete in 2000, so, _please_, let's not go > for monikers that have long ran their course when introducing new files > and function calls. And yes, I know that NT is still being used by > Windows and Microsoft internally, but that's only because of legacy... > which we're absolutely not being bound to here. Or do we anticipate that > we'll have to support a different Windows architecture than NT in the > future? I've had an interesting range of reactions to your objection. Let me try to present some other ways to think about it. It's true that Microsoft stopped using Windows NT in its marketing material as of Windows XP (the Windows 2000 tag line was the redundant phrase "Built on NT Techology"), but it's also true that the system we now generically call Windows uses the NT kernel, essentially unchanged in its fundamental design from its 1989 origins. That term is not inaccurate. This is not terribly different from Linux, where we have marketing terms like Ubuntu and Fedora and Mint that are used to describe operating systems that use the Linux kernel, which is only slightly younger than the NT kernel. Microsoft would like us to think that there is only One True Windows, but that's not quite true yet, as long as CE still lives. In my personal opinion, there's nothing wrong with having _nt_ in the file names, but I would just as soon remove "_nt_" from the function names. -- Tim Roberts, ti...@pr... Providenza & Boekelheide, Inc. |