Return to BSD News archive
Xref: sserve comp.unix.solaris:488 comp.unix.bsd:7788 Path: sserve!manuel.anu.edu.au!munnari.oz.au!spool.mu.edu!uwm.edu!cs.utexas.edu!usc!usc!not-for-mail From: merlin@neuro.usc.edu (merlin) Newsgroups: comp.unix.solaris,comp.unix.bsd Subject: Re: Solaris 1.1 vs. Solaris 2.0 (BSD vs AT&T) Date: 15 Nov 1992 02:11:58 -0800 Organization: University of Southern California, Los Angeles, CA Lines: 30 Distribution: inet Message-ID: <1e57peINNg87@neuro.usc.edu> References: <BxLz6x.EL7@cs.uiuc.edu> <1992Nov13.232053.7061@sjsumcs.sjsu.edu> <VIXIE.92Nov14194825@cognition.pa.dec.com> NNTP-Posting-Host: neuro.usc.edu In article <VIXIE.92Nov14194825@cognition.pa.dec.com> vixie@pa.dec.com (Paul A Vixie) writes: >You mean I'll be back to the 14-character file name limit soon? Or that >I'll have to convert all my code from sockets to STREAMS? Or that I'll >have to run a 200-kilobyte shell script to add accounts to my computer >soon? Or that I'll have to punt X Windows in favor of a DMD5620 or VT100? >Or that I'll have to log into ksh or sh since will be removed from my disk? On my SCO SYSV/386 3.2r2.0 ODT 1.1 system we have BSD sockets, menu driven account creation, and X11R4 server and clients. ODT 2.0 adds symbolic links and much longer filenames. What will be removed from your disk? We have sh, csh, ksh -- and probably anything else you would want. >Face it, dude, AT&T doesn't know UNIX from a hole in their ass. USL is >even less clueful. System V UNIX is dead. The market opened their eyes >and told them "you can't possibly be serious!" and they adopted BSD as >the only way to keep selling licenses. POSIX won the interface battle, >not SVID. BSD won the users over. If you think a new user would take >System V.[01234] seriously as a competitor to Windows/NT or BSD, you are >totally out of your freaking mind. Windows/NT is likely to have BSD and AT&T/USL UNIX for lunch. It supposedly offers POSIX, Win16, Win32, SMP, networking, and a whole truckload of other stuff on a single optical disk for easy installation and operation. I have not seen an X11R5 product announcement -- but it can't be ffar from release. I suspect most desktop systems will be running Windows/NT by the end of 1993 -- and with rare exceptions i386/i486 based SYSV systems will bite the dust. The large scale mainframe market may still be up for grabs -- but Windows/NT as a transportable scalable architecture (provided the claims are met) will probably make a very attractive cross spectrum operating system product.