home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!portal!cup.portal.com!ts
- From: ts@cup.portal.com (Tim W Smith)
- Newsgroups: comp.os.os2.programmer
- Subject: Re: Win NT Dev, Conf. Report
- Message-ID: <62521@cup.portal.com>
- Date: Tue, 21 Jul 92 07:33:55 PDT
- Organization: The Portal System (TM)
- References: <06!m_td.rcain@netcom.com>
- <pathak-130792094456@virtual.mitre.org>
- Lines: 24
-
- > My biggest concern here has been the stability of the kernel. While the
- > beta kernel looks good, you really don't know how well it works. Before
- > somebody posts something about how it is a microkernel and thus small and
- > easy to write well, let me point out that both NeXT and OSF spend a huge
- > amount of time testing and debugging the Mach microkernel. From what I
- > understand, the Mach kernel is now very stable but if it took time and
- > people using in commercial situtations before all the bugs were found and
- > taken out.
-
- Yeah, but compared to the number of people who develop for Microsoft
- OSes, the number of Mach developers is insignificant. If Microsoft
- can get a significant fraction of their developers to participate in
- an NT beta, they should be able to get kernel stability before a general
- release with no problem (assuming that any lack of kernel stability is
- due to implementation bugs rather than design flaws).
-
- They can do what Apple did with System 7 for the Mac. Apple made sure
- that alphas and betas of System 7 where widespread in the developer
- community. They had many thousands of qualified alpha and beta testers.
-
- A system like Mach is not going to get this kind of testing until it
- gets released.
-
- --Tim Smith
-