llxxkk@mixcom.UUCP (Adam Costello) (07/20/90)
I recently told a friend that the Amiga 3000 would soon be running AT&T UNIX System V Release 4.0, and he said that I would be much better off with a BSD release. I asked why, and he gave me the following summary: quote What is wrong with Sys V: No finger No who No what ps works funny (and not nearly as well) echo is hacked the csh is not supported worth dog dooky stty has not half the options shell scripts execute with the Bourne Shell no matter what the #!<interpreter> line for scripts is not supported the csh is crippled (very poor job control) the only alternative to the csh is ksh, a worthless hack on sh BSD network niceties are not supported (like telnet) In a nut shell, it sucks. (Why else does Sun ship BSD?) end of quote My question is, is he accurate, and if so, do his criticisms apply to Release 4.0? Please send replies via email to llxxkk@mixcom.UUCP, or post to comp.sys.amiga, since I don't normally subscribe to this newsgroup. Your help is much appreciated. Thanks, AMC
guy@auspex.auspex.com (Guy Harris) (07/21/90)
(More generic UNIX than AMIX-specific. Presumably, Commodore won't blow off any of the S5R4 stuff the guy asked for....) >I recently told a friend that the Amiga 3000 would soon be running AT&T UNIX >System V Release 4.0, and he said that I would be much better off with a BSD >release. I asked why, and he gave me the following summary: Too bad he didn't check out S5R4.0 first.... >What is wrong with Sys V: > > No finger See finger(1) in the S5R4 User's Reference Manual. > No who Rubbish - "who" dates back to ancient times, and S5 has always had it. Perhaps there's some specific *feature* of "who" to which he's referring? > No what Rubbish - "what" originally came with SCCS, back in PWB/UNIX 1.0 days or so; Berkeley created their own, not from AT&T source (not their own SCCS, they use AT&T-derived source for that, although since it's not part of 32V they don't distribute it with BSD). > ps works funny (and not nearly as well) "ps" does have different flags, if that's what he means by "works funny"; dunno what he means by "not nearly as well". One thing *I* prefer about the S5 "ps" is that, by default, it shows the processes running on your terminal, not the processes running with your user ID; I rarely care about the latter, and, if not asking for all the processes, usually care about the former. > the csh is not supported worth dog dooky The "csh" in S5R4 is derived from the SunOS 4.1 one, which is derived from the 4.3BSD one; it's not the crappy ancient one that appears to come with older S5 releases. > stty has not half the options "stty" has more options in S5R4 than in BSD, including all the nice user interface stuff such as echoing control characters as ^x rather than as themselves. It doesn't have "ltilde", but that's a bit of a crock anyway. > shell scripts execute with the Bourne Shell no matter what > the #!<interpreter> line for scripts is not supported S5R4 supports "#!". This obviously lets you have your scripts be run by the Bourne shell, the C shell, the Korn shell, the Bourne-again shell, "awk", "sed", "perl", "vi", "make", ... if you have them. > the csh is crippled (very poor job control) The S5R4 "csh" - as well as the S5R4 "ksh", *and* the S5R4 Bourne shell if you run it as "jsh" - supports job control. > the only alternative to the csh is ksh, a worthless hack on sh Frankly, I think "csh" is a nearly-worthless hack on the V6 shell, and wish I had the Korn shell (no, "bash" won't cut it until they implement ^O just like in EMACS mode in "ksh"), but to each their own.... > BSD network niceties are not supported (like telnet) See ftp(1), telnet(1), rsh(1), rlogin(1), rcp(1), etc. in the S5R4 User's Reference Manual. >In a nut shell, it sucks. (Why else does Sun ship BSD?) Sun doesn't ship BSD, Sun ships SunOS. SunOS probably has more BSD stuff in it than S5 stuff, but it has a fair bit of both, as well as a fair bit of stuff that comes from neither (much of which is in S5R4). Yes, I'd miss the BSD stuff if it weren't there; I'd miss a lot of the S5 stuff, too. Other vendors ship S5 systems (or systems billed as such) with BSD stuff added in; AT&T, as of S5R4, is one such vendor (and even as of some earlier systems, although with less BSD stuff). >My question is, is he accurate, Only in some cases. >and if so, do his criticisms apply to Release 4.0? Few, if any.
dhesi%cirrusl@oliveb.ATC.olivetti.com (Rahul Dhesi) (07/22/90)
For about two years now, every time somebody has mentioned the System V versus BSD battle (which was really lost and won years ago), somebody else has invariably pointed to the not-yet-here-but-real-soon-now SVR4 as the panacea. But most of the general public that's stuck with System V has been really stuck with SVR3 (or in some cases SVR2, and in one square-case-with-a-floppy, with SVR1). By the time SVR4 becomes widely available, 4.4BSD and commercial OSs derived from it will be becoming available and the debate will begin again....And this time around, presumably the as-yet-nonexistent SVR5 will be the panacea. -- Rahul Dhesi <dhesi%cirrusl@oliveb.ATC.olivetti.com> UUCP: oliveb!cirrusl!dhesi
jemearl@ms.uky.edu (Jemearl T. Smith) (07/23/90)
From the two weeks of classes I took in New Jersy on the new AT&T
WGS product line, System VR4 was discussed. It is a hybrid of System
V (all of them), Xenix and BSD. It will be compatable with all of
the old system V's, Xenix and BSD, so binary portablily is possible.
All of this came about when Unix International was getting tired of the
"Major Unix War" between AT&T and BSD. So, these two teams got together
and made UNIX System 5 Relese 4.
I, personally, can't wait.
--
-- Jemearl T. Smith // Ondur Ildoch'i (SCA)
-- Archiving Dude, and Amiga owner. \X/ Dragonsmark
-- jemearl@ms.uky.edu jemearl@UKMA.BITNET !ukma!jemearl {UUNET}
-- I'd be ok if only my girlfriend didn't call me a yuppie mongol ag@cbmvax.commodore.com (Keith Gabryelski) (07/24/90)
In article <3698@auspex.auspex.com> guy@auspex.auspex.com (Guy Harris) writes: >(More generic UNIX than AMIX-specific. Presumably, Commodore won't >blow off any of the S5R4 stuff the guy asked for....) Commodore will not be "blowing off the S5R4 stuff the guy asked for". Pax, Keith
ndjc@hobbit.UUCP (Nick Crossley) (07/24/90)
In article <2059@cirrusl.UUCP> dhesi%cirrusl@oliveb.ATC.olivetti.com (Rahul Dhesi) writes: >For about two years now, every time somebody has mentioned the System V >versus BSD battle (which was really lost and won years ago), somebody >else has invariably pointed to the not-yet-here-but-real-soon-now SVR4 >as the panacea. V.4 is not a myth, it is available NOW, and has been since early this year (earlier in Europe than in the US!). ICL has already shipped over 300 of our DRS6000 SPARC based systems, running Unix V.4. I do not believe we are the only manufacturer now doing so, either. disclaimer: I work for ICL! Nick Crossley, ICL NA, 9801 Muirlands, Irvine, CA 92718-2521, USA 714-458-7282 uunet!ccicpg!ndjc / ndjc@ccicpg.UUCP -- <<< standard disclaimers >>> Nick Crossley, ICL NA, 9801 Muirlands, Irvine, CA 92718-2521, USA 714-458-7282 uunet!ccicpg!ndjc / ndjc@ccicpg.UUCP
peter@ficc.ferranti.com (Peter da Silva) (07/24/90)
In article <2059@cirrusl.UUCP> dhesi%cirrusl@oliveb.ATC.olivetti.com (Rahul Dhesi) writes: > For about two years now, every time somebody has mentioned the System V > versus BSD battle (which was really lost and won years ago),... Yep. And the winners are manufacturers of memory devices, and the losers are the users... If automobiles had developed as fast as computer software, today a Rolls Royce would cost 30 million dollars, require thousands of gallons to the mile, go 2 miles per hour, but be able to carry a space shuttle on the roof rack. Just in case. -- Peter da Silva. `-_-' +1 713 274 5180. 'U` <peter@ficc.ferranti.com>
gwyn@smoke.BRL.MIL (Doug Gwyn) (07/24/90)
In article <179@mixcom.UUCP> llxxkk@mixcom.UUCP (Adam Costello) writes:
-My question is, is he accurate, and if so, do his criticisms apply to Release>release. I asked why, and he gave me the following summary:
-4.0?
He is almost totally wrong.gwyn@smoke.BRL.MIL (Doug Gwyn) (07/24/90)
In article <2059@cirrusl.UUCP> dhesi%cirrusl@oliveb.ATC.olivetti.com (Rahul Dhesi) writes: >By the time SVR4 becomes widely available, 4.4BSD and commercial OSs >derived from it will be becoming available and the debate will begin >again... I don't see any signs at all that commercial vendors are going to distribute systems based on 4.4BSD rather than SVR4. Some may ship Mach-based systems, but they will be at a competitive disadvantage.
peter@ficc.ferranti.com (Peter da Silva) (07/24/90)
In article <13399@smoke.BRL.MIL> gwyn@smoke.BRL.MIL (Doug Gwyn) writes: > I don't see any signs at all that commercial vendors are going to > distribute systems based on 4.4BSD rather than SVR4. Some may ship > Mach-based systems, but they will be at a competitive disadvantage. Why, if it can be made to conform to the SVID and applicable ABIs? The additional functionality and performance available through the use of a modern kernel instead of a swiss-army-knife coroutine-based hack should more than outweigh the extra development costs. If they can do it without a UNIX source licence it may even be cheaper. And I'd *love* to be able to *not* load all that BSD and Sys V compatibility crud if I don't want to. -- Peter da Silva. `-_-' +1 713 274 5180. 'U` <peter@ficc.ferranti.com>
guy@auspex.auspex.com (Guy Harris) (07/25/90)
>From the two weeks of classes I took in New Jersy on the new AT&T >WGS product line, System VR4 was discussed. It is a hybrid of System >V (all of them), Xenix and BSD. It will be compatable with all of >the old system V's, Xenix and BSD, so binary portablily is possible. Well, err, umm, binary portability is possible within limits, in the case of BSD. Binary portability from BSD to S5R4 might be possible with enough work, but you won't get it for free. The 32-bit UIDs and GIDs in S5R4's BSDish "stat" structure get in the way, for example. If a vendor is, for example, offering a BSD-based system and switches to S5R4, they'll have to do some work to provide binary compatibility. Lots of programs written for BSD will be portable in *source* form to S5R4; some more will be portable if the vendor includes the BSD compatibility package.
guy@auspex.auspex.com (Guy Harris) (07/25/90)
>I don't see any signs at all that commercial vendors are going to >distribute systems based on 4.4BSD rather than SVR4. Some may ship >Mach-based systems, but they will be at a competitive disadvantage. Well, it'll depend on how they use the Mach base. If they implement an S5R4 interface atop it, incompatibility probably won't provide any competitive disadvantage (especially if they make it binary-compatible as well).
mikep@dirty.dsg.ti.com (Michael A. Petonic) (07/26/90)
In article <2059@cirrusl.UUCP> dhesi%cirrusl@oliveb.ATC.olivetti.com (Rahul Dhesi) writes: >For about two years now, every time somebody has mentioned the System V >versus BSD battle (which was really lost and won years ago), somebody >else has invariably pointed to the not-yet-here-but-real-soon-now SVR4 >as the panacea. But most of the general public that's stuck with >System V has been really stuck with SVR3 (or in some cases SVR2, and in >one square-case-with-a-floppy, with SVR1). > >By the time SVR4 becomes widely available, 4.4BSD and commercial OSs >derived from it will be becoming available and the debate will begin >again....And this time around, presumably the as-yet-nonexistent SVR5 >will be the panacea. So -- what's your point? I thought this was comp.unix.*questions*, not comp.unix.religious.wars. -MikeP
lars@iclswe.icl.se (Lars Tunkrans) (07/28/90)
ndjc@hobbit.UUCP (Nick Crossley) writes: >In article <2059@cirrusl.UUCP> dhesi%cirrusl@oliveb.ATC.olivetti.com (Rahul Dhesi) writes: >>For about two years now, every time somebody has mentioned the System V >>versus BSD battle (which was really lost and won years ago), somebody >>else has invariably pointed to the not-yet-here-but-real-soon-now SVR4 >>as the panacea. >V.4 is not a myth, it is available NOW, and has been since early this year >(earlier in Europe than in the US!). ICL has already shipped over 300 of >our DRS6000 SPARC based systems, running Unix V.4. I do not believe we are >the only manufacturer now doing so, either. >disclaimer: I work for ICL! >Nick Crossley, ICL NA, 9801 Muirlands, Irvine, CA 92718-2521, USA 714-458-7282 >uunet!ccicpg!ndjc / ndjc@ccicpg.UUCP >-- Nick is just slightly wrong we have delivered more than 700 systems , I know because I installed and commisioned serial 703 last week ! Lars. -- Lars Tunkrans Phone +46 (0)76096368. | The ICL DRS6000 SPARC system is still DRS Systems Support. | the only off the shelf deliverable UUCP: uunet!mcsun!sunic!iclswe!lars | Unix System V Release 4.0 system. DOMAIN Address : lars@iclswe.icl.se |
ndjc@hobbit.UUCP (Nick Crossley) (07/31/90)
In article <1990Jul27.203756.18208@iclswe.icl.se> lars@iclswe.icl.se (Lars Tunkrans) writes: >ndjc@hobbit.UUCP (Nick Crossley) writes: >>ICL has already shipped over 300 of our DRS6000 SPARC based systems, >>running Unix V.4. > >Nick is just slightly wrong we have delivered more than 700 systems , I know >because I installed and commisioned serial 703 last week ! > My figures might nave been slightly out of date, particularly in respect to the European sales. However, but quite a few of those systems were for internal use (we have about 40 or so in our development labs here. I was referring strictly to customer shipments, as the original poster was querying the PUBLIC availability of V.4. -- <<< standard disclaimers >>> Nick Crossley, ICL NA, 9801 Muirlands, Irvine, CA 92718-2521, USA 714-458-7282 uunet!ccicpg!ndjc / ndjc@ccicpg.UUCP