home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.os.ms-windows.programmer.win32
- Path: sparky!uunet!utcsri!torn!cunews!csi.uottawa.ca!news
- From: duwors@csi.uottawa.ca (Robert J. Du Wors)
- Subject: Re: Looking for rshd, telnetd, ftpd etc.
- Message-ID: <1992Sep3.180007.25918@csi.uottawa.ca>
- Sender: news@csi.uottawa.ca
- Nntp-Posting-Host: prgc
- Organization: Dept. of Computer Science, University of Ottawa
- References: <1992Aug28.000903.22215@microsoft.com> <1992Sep01.191707.28916@news.mentorg.com> <1992Sep02.211813.14491@microsoft.com>
- Date: Thu, 3 Sep 92 18:00:07 GMT
- Lines: 39
-
- In article <1992Sep02.211813.14491@microsoft.com> michaelw@microsoft.com (Michael Winser) writes:
- >In article <1992Sep01.191707.28916@news.mentorg.com> pbrooks@mentorg.com (Phil Brooks) writes:
- >
- >Once again, why stick with the unix model? Services are explicitly supported
- >by Windows NT. When started they take on the ACE of the user that installed
- >them. No login required. I'm pretty sure that you can start these services
- >via RPC. It would be trivial to write a start server daemon (for RPC, telnet
- >or whatever).
-
- Why stick with the "UNIX" model? Because it works, and works well. Hype
- aside, why thow out basic ASCII streams as a fundamental network primitive?
- (i.e. virtual circuit model with call setup and tear down, more commonly known
- in the OS community as "login"/"logout"). NT Servers are going to get awfully
- lonely in mixed environmnets. Perhaps as lonely as Apple "servers", if that
- isn't an outright oxymoron.
-
- It is along way from a kludged individual solution, to a clean and supported
- interface. Consider also that the most successful example of data flow
- programming in the world to day is shell scripts and pipes, including the true
- ease of integration via rsh. It can even be done under VMS. Your might
- change the script language, you might have many of them, but the basic
- requirements transcends all of them.
-
- Finally, the Internet is not an accident, a foolish standard set by committee,
- nor an attempt at market dominance...and it has survived and prospered on this
- "UNIX" model (sic). Imagine writing your own client program for every different
- server on the net...of course that is exactly the case with Compu$erve. All
- this single source stuff, and where is the improvement...
-
- I think that a new OS needs a little breathing room but that doesn't mean its
- advocates need to deny what they will be forced to provide by the market place.
- (You can bet your old friends at IBM once vowed never to violate the purity of
- MVS/SNA with TCP/IP and its upper level services).
-
- Rob D.
-
- -----------------------
- duwors@csi.uottawa.ca aka Robert Du Wors Tel:613.831.1930 Fax:613.831.0272
- 141 Country Lane, Kanata, Ontario, Canada K2L 1J6
-