chapman@eris.BERKELEY.EDU (Brent Chapman) (01/06/88)
In article <11106@brl-adm.ARPA> GUTHERY%ASC%sdr.slb.com@RELAY.CS.NET (guthery%asc@sdr.slb.com) writes: >OS/2 does things UNIX doesn't. Like what? How about some examples, for those of us with little experience with OS/2 but lots of distrust for IBM, MicroSoft, and marketroids in general? >It's benefited from 15 years of learning and research. And UNIX hasn't? Come on... UNIX has improved a hell of a lot in just the few years that I've been using it.. >SIGOPS shouldn't close its doors. What's the big deal? What is "SIGOPS"? >Sun tried for both threads and dynamic linking and blew it. Can you clarify that? I run an all-Sun shop, so I'm mighty interested if you can back your claims. (On the other hand, you may be just blowing smoke...) >There's a message there. Where? All I see are a few grumbling generalities, without any explanation to back them up... -Brent -- Brent Chapman Capital Market Technology, Inc. Senior Programmer/Analyst 1995 University Ave., Suite 390 {lll-tis,ucbvax!cogsci}!capmkt!brent Berkeley, CA 94704 capmkt!brent@{lll-tis.arpa,cogsci.berkeley.edu} Phone: 415/540-6400
GUTHERY%ASC%sdr.slb.com@RELAY.CS.NET (guthery%asc@sdr.slb.com) (01/07/88)
1) Here are some O/S goodies that stock OS/2 (now playing on my desk) has that stock Unix (also playing on my desk) doesn't have: - built-in light weight processes blended compatibly with heavy weight processes - runtime dynamic linking and demand loading - shared global memory segments - file locking by byte region - standard system calls from drivers - periodic signals - systemwide semaphores - file write-though - system trace 2) Microsoft/IBM marketing hype seems to bother some folks. I will wager a chocolate Dunkin' Donut that the number of dollars spent on hyping Unix divided by the number of years Unix has been out is GREATER than the number of dollars spent hyping OS/2 divided by the (fractional) number years its been out. I think it's been the Year of Unix since about 1982, hasn't it? (Yawn) AT&T by itself has spent far more hyping Unix than Microsoft and IBM put together have spent hyping OS/2 AND DOS. Tell me again about hype. 3) Kernel hacking is not research. A Unix wizard who doesn't know what SIGOPS is loses five hit points. Do the wizards sense extinction? Cheers, Scott
edw@IUS1.CS.CMU.EDU (Eddie Wyatt) (01/08/88)
Please specify what flavor of Unix you are talking about because.... In article <11156@brl-adm.ARPA>, GUTHERY%ASC%sdr.slb.com@RELAY.CS.NET (guthery%asc@sdr.slb.com) writes: > 1) Here are some O/S goodies that stock OS/2 (now playing on my desk) > has that stock Unix (also playing on my desk) doesn't have: > - built-in light weight processes blended compatibly > with heavy weight processes Mach a Unix kernal (4.2 BSD compadible) has threads and tasks light weight and heavy weight process respectively. > - runtime dynamic linking and demand loading Is posible under BSD 4.2 Unix - I don't know how its done but we have someone around here maintaining a load at run time library. > - shared global memory segments Sys V has shared memory segments. So does Sun 3.2 and later additions after merging BSD and Sys V. > - file locking by byte region Sys V file locking. > - standard system calls from drivers That's an advantage? > - periodic signals BDS SIGALRM > - systemwide semaphores SYS V maybe I'm not sure. > - file write-though Not sure what you mean. > - system trace ptrace ? I'm not condemning OS/2 never played with it, just stating some of the features you attribute to only OS/2 are in a number of flavors of Unix. -- Eddie Wyatt e-mail: edw@ius1.cs.cmu.edu
jpn@teddy.UUCP (John P. Nelson) (01/08/88)
>3) Kernel hacking is not research. A Unix wizard who doesn't know what >SIGOPS is loses five hit points. Do the wizards sense extinction? I just lost 5 hit points. Can anyone enlighten me? I would have sent mail, but the return address had too many '%', '@', and '!' characters in it.
dave@sdeggo.UUCP (David L. Smith) (01/08/88)
In article <11156@brl-adm.ARPA>, GUTHERY%ASC%sdr.slb.com@RELAY.CS.NET (guthery%asc@sdr.slb.com) writes: > 1) Here are some O/S goodies that stock OS/2 (now playing on my desk) > has that stock Unix (also playing on my desk) doesn't have: > - built-in light weight processes blended compatibly > with heavy weight processes > - runtime dynamic linking and demand loading > - shared global memory segments > - file locking by byte region > - standard system calls from drivers > - periodic signals > - systemwide semaphores > - file write-though > - system trace Which version of Unix are you running? Both System V and BSD have file locking by byte regions through the fcntl() and lockf() calls respectively. (Ok, ok, so they're advisory in many implementations.) Are "shared global memory segments" equivalent to the Sys V shared memory or is there a nuance I'm missing here? What's the difference between a "systemwide semaphore" and Sys V semaphores? Dynamic linking is supported on several machines through shared libraries, however these may not be as nice as OS/2's Dynalinks. Demand loading is done by BSD 4.x systems and Sys V.3, is it not? OS/2 certainly does things that Unix does not...yet. However, it is unlikely to take over the market if it is, as I have heard, a hardware specific OS, aimed mainly at running on 80286's. And it's probably not going to run on my '286, except as a novelty, because I can't take the program I write under OS/2 which uses all the nifty OS/2 features and run it on the machine at work, as I can with the programs off my Microport Sys V. -- David L. Smith {sdcsvax!jack,ihnp4!jack, hp-sdd!crash, pyramid}!sdeggo!dave sdeggo!dave@amos.ling.edu Sinners can repent, but stupid is forever.
chris@mimsy.UUCP (Chris Torek) (01/08/88)
In article <11156@brl-adm.ARPA> GUTHERY%ASC%sdr.slb.com@RELAY.CS.NET (guthery%asc@sdr.slb.com) writes: >1) Here are some O/S goodies that stock OS/2 (now playing on my desk) >has that stock Unix (also playing on my desk) doesn't have: > - built-in light weight processes blended compatibly > with heavy weight processes > - runtime dynamic linking and demand loading > - shared global memory segments > - file locking by byte region > - standard system calls from drivers > - periodic signals SunOS 4.0 (of course it is not *out* yet) has all but two of those. The two are lightweight processes (hard to do right, as the Mach group at CMU discovered, because no one agrees as to what is `right') and system calls from drivers. The latter is a horrible thing to do to a kernel (my opinion, but shared by a number of friends). > - systemwide semaphores > - file write-though > - system trace As for these, I am not sure what is meant. SunOS 4.0 has SysV semaphores and mapped files. If you have 4BSD-based *source* (such as SunOS 3.x---not having seen 4.0 source I am not sure it is still there, though I imagine so), there is a system call trace facility, but it is very primitive. On the other hand, source level debuggers can trace system calls from user programs. (Having kernel source is usually required because most vendors compile without -DSYSCALLTRACE.) -- In-Real-Life: Chris Torek, Univ of MD Comp Sci Dept (+1 301 454 7690) Domain: chris@mimsy.umd.edu Path: uunet!mimsy!chris
chris@mimsy.UUCP (Chris Torek) (01/08/88)
>>3) Kernel hacking is not research. A Unix wizard who doesn't know what >>SIGOPS is loses five hit points. Do the wizards sense extinction? In article <4558@teddy.UUCP> jpn@teddy.UUCP (John P. Nelson) writes: >I just lost 5 hit points. Can anyone enlighten me? Kernel hacking *can* be research; it just has to be done *right*. Actually, sufficiently *wrong* works well too :-) . But to spoil the fun, SIGOPS <=> Special Interest Group on OPerating Systems, one of many ACM (Association for [not `of' ... yet?] Computing Machinery) SIGs. -- In-Real-Life: Chris Torek, Univ of MD Comp Sci Dept (+1 301 454 7690) Domain: chris@mimsy.umd.edu Path: uunet!mimsy!chris
chapman@eris.BERKELEY.EDU (Brent Chapman) (01/08/88)
In article <11156@brl-adm.ARPA> GUTHERY%ASC%sdr.slb.com@RELAY.CS.NET (guthery%asc@sdr.slb.com) writes: >3) Kernel hacking is not research. A Unix wizard who doesn't know what >SIGOPS is loses five hit points. Do the wizards sense extinction? I neither claim to be a kernel hacker or a Unix wizard (expert maybe, but I'm not a wizard yet). The only SIGOPS I know of is ACM's Special Interest Group on Operating Systems; if that's what you're referring to, I still don't understand your cryptic comment about "SIGOPS closing its doors". -Brent -- Brent Chapman Capital Market Technology, Inc. Senior Programmer/Analyst 1995 University Ave., Suite 390 {lll-tis,ucbvax!cogsci}!capmkt!brent Berkeley, CA 94704 capmkt!brent@{lll-tis.arpa,cogsci.berkeley.edu} Phone: 415/540-6400
gwyn@brl-smoke.ARPA (Doug Gwyn ) (01/08/88)
In article <11156@brl-adm.ARPA> GUTHERY%ASC%sdr.slb.com@RELAY.CS.NET (guthery%asc@sdr.slb.com) writes: >1) Here are some O/S goodies that stock OS/2 (now playing on my desk) >has that stock Unix (also playing on my desk) doesn't have: UNIX has about half of these. Perhaps you're confusing UNIX with 4BSD?
rwa@auvax.UUCP (Ross Alexander) (01/09/88)
In article <11156@brl-adm.ARPA>, GUTHERY%ASC%sdr.slb.com@RELAY.CS.NET (guthery%asc@sdr.slb.com) writes: > 1) Here are some O/S goodies that stock OS/2 (now playing on my desk) > has that stock Unix (also playing on my desk) doesn't have: ----- Which one, prithee? Inqiring minds want to know! Ross Alexander @ Athabasca University alberta!auvax!rwa
allbery@ncoast.UUCP (Brandon Allbery) (01/10/88)
[We haven't seen the line-eater since we sent Minnifield and Dixon after it!] Eddie Wyatt does a good job of defending UNIX vs. OS/2; I'll add a few blows... As quoted from <603@PT.CS.CMU.EDU> by edw@IUS1.CS.CMU.EDU (Eddie Wyatt): +--------------- | In article <11156@brl-adm.ARPA>, GUTHERY%ASC%sdr.slb.com@RELAY.CS.NET (guthery%asc@sdr.slb.com) writes: | > 1) Here are some O/S goodies that stock OS/2 (now playing on my desk) | > has that stock Unix (also playing on my desk) doesn't have: | | > - systemwide semaphores | | SYS V maybe I'm not sure. +--------------- System V definitely. +--------------- | > - file write-though | | Not sure what you mean. +--------------- If this means "synchronous writes", i.e. a write() writes to the disk instead of a buffer, try System V open(filename, O_WRONLY|O_SYNC). +--------------- | > - system trace | ptrace ? +--------------- Tracing the system? I think (not sure yet) SVR3's streams logging may be able to do this. I'm still waiting for full manuals... but if it doesn't I suspect it could be added easily. I'm not sure I see why it's needed, though. I've done this under microcomputer OSes, but this is usually to detect a program corrupting system memory (which can't happen under UNIX -- is this the reason it is provided in OS/2?) Anyway, looks like between System V, BSD, and Mach, OS/2 is covered. After the Sun/Microsoft/AT&T merge, UNIX will be all set. I just wish they'd hurry it up a bit! -- Brandon S. Allbery, moderator of comp.sources.misc {well!hoptoad,uunet!hnsurg3,cbosgd,sun!mandrill}!ncoast!allbery WARNING!!! Site "necntc" apparently is no longer a mail feed for ncoast! PS/2: Half a computer. OS/2: Half an operating system for half a computer.
ram%shukra@Sun.COM (Renu Raman, Sun Microsystems) (01/11/88)
In article <7067@ncoast.UUCP> allbery@ncoast.UUCP (Brandon Allbery) writes: >[We haven't seen the line-eater since we sent Minnifield and Dixon after it!] > The biggest plus is portability. 80X86 is not the world. Not everybody like apples and not everbody like oranges. If and when OS/2 runs on my desktop machine (you can guess the processor) as well as any processor that ranges from X86 to Cray's to IBM's biggies with all the features that you mention, then it deserves to be compared with UNIX. two questions: Looks like the features in OS/2 will make VM/CMS and other IBMers drool. Will IBM provide OS/2 on their mainframes? Would you get source if you want to? --------------------- Renukanthan Raman ARPA:ram@sun.com Sun Microsystems UUCP:{ucbvax,seismo,hplabs}!sun!ram M/S 5-40, 2500 Garcia Avenue, Mt. View, CA 94043
jfh@killer.UUCP (The Beach Bum) (01/11/88)
In article <448@minya.UUCP>, jc@minya.UUCP (John Chambers) writes: > In article <11156@brl-adm.ARPA>, GUTHERY%ASC%sdr.slb.com@RELAY.CS.NET (guthery%asc@sdr.slb.com) writes: > I'll just point out that every Unix I've ever seen has > systemwide semaphores. They just aren't called that. Try: > if ((l = creat("/usr/spool/locks/<somename>",0)) { > ... > [Critical section] > ... > unlink("/usr/spool/locks/<somename>"); > } else { > [Didn't get semaphore] > } > It worked on BRL Version 6, and it still works, even on SunOS. I understand there is an old race in the V7 and System III kernels that causes this to not work reliably. Also, it doesn't work at all if root executes the code as root will just re-create the file AND return success to boot. [ isn't that special? ;-) ] For System V, use the three argument form of open(2) and OR in the O_EXCLUDE flag. This insures ;-) ;-) that the file does not exist before trying the open. Label me paranoid, but, I prefer to stick an access() in the if() before the open(). Just in case ... - John. -- John F. Haugh II SNAIL: HECI Exploration Co. Inc. UUCP: ...!ihnp4!killer!jfh 11910 Greenville Ave, Suite 600 "Don't Have an Oil Well? ... Dallas, TX. 75243 ... Then Buy One!" (214) 231-0993 Ext 260
dfields@ccvaxa.UUCP (01/12/88)
/* Written 8:34 pm Jan 7, 1988 by chris@mimsy.UUCP in ccvaxa:comp.unix.wizards */ >In article <11156@brl-adm.ARPA> GUTHERY%ASC%sdr.slb.com@RELAY.CS.NET >(guthery%asc@sdr.slb.com) writes: >> - file write-though >> - system trace > >As for these, I am not sure what is meant. SunOS 4.0 has SysV >semaphores and mapped files. If you have 4BSD-based *source* (such >as SunOS 3.x---not having seen 4.0 source I am not sure it is still >there, though I imagine so), there is a system call trace facility, >but it is very primitive. On the other hand, source level debuggers >can trace system calls from user programs. (Having kernel source >is usually required because most vendors compile without -DSYSCALLTRACE.) >-- >In-Real-Life: Chris Torek, Univ of MD Comp Sci Dept (+1 301 454 7690) >Domain: chris@mimsy.umd.edu Path: uunet!mimsy!chris It's been a year and a half since I've worked with it but I believe "file write-through" is marketing-ese for no file system cache. The system call trace was going to be used to debug crashes. There's a utility which will print the last XXX trace points in the kernel. It wasn't very pretty the last time I saw it but.... The things that bothered me about OS/2 were that kernel written in assembler because some people were worried about efficiency (some of the code was very ineffcient though) and that the utilites and file system were the same brain damaged ones as on MS-DOS. They do have the hooks to add different file systems and you can buy utilities from third-parties (soon at least) but I can't recommend it. There are some nice improvements over MS-DOS but I (as a developer) would rather use U*X and would advise my friends who just want to run useful software on a home machine to buy a Mac. The contents of this note are, of course, my and only my responsibility. Dave Fields. Gould CSD-Urbana. USEnet: ihnp4!uiucdcs!ccvaxa!dfields 1101 E. University, Urbana, IL 61801 ARPAnet: dfields@gswd-vms.arpa
daveb@llama.rtech.UUCP (Dave Brower) (01/12/88)
In article <10091@mimsy.UUCP> chris@mimsy.UUCP (Chris Torek) writes: >In article <11156@brl-adm.ARPA> GUTHERY%ASC%sdr.slb.com@RELAY.CS.NET >> - file write-though > >As for these, I am not sure what is meant... A critical issue to database types: Making sure the contents of a write actually hit the disk. Fsync sort of does the job on BSD (at too high a price), as does O_SYNC on some Sytem V's. Unfortunately, O_SYNC is a bit too zealous, as it forces a write of both the data block and the inode (for mod time). The latter is often overkill, and might better be a separate operation. You don't always care if the mod time is right, and not want to pay the cost of another real disk i/o. I don't know if OS/2 does _the_right_thing_ in this regard, thogh it seems to have some flavor of the feature. Maybe it's like VMS: no OS cache means guaranteed writethrough :-) -dB "I don't care what you say, as long as you spell my name right." {amdahl, cbosgd, mtxinu, ptsfa, sun}!rtech!daveb daveb@rtech.uucp
ericb@athertn.Atherton.COM (Eric Black) (01/12/88)
In article <448@minya.UUCP> jc@minya.UUCP (John Chambers) writes: >In article <11156@brl-adm.ARPA>, GUTHERY%ASC%sdr.slb.com@RELAY.CS.NET (guthery%asc@sdr.slb.com) writes: >> - systemwide semaphores >I'll let the other wizards hack apart the rest of the items on >the list; I'll just point out that every Unix I've ever seen has >systemwide semaphores. They just aren't called that. Try: > if ((l = creat("/usr/spool/locks/<somename>",0)) { > ... > [Critical section] > ... > unlink("/usr/spool/locks/<somename>"); > } else { > [Didn't get semaphore] > } >It worked on BRL Version 6, and it still works, even on SunOS. ^^^^^ Careful in an NFS environment! It doesn't work (where "work" is defined as "guarantees exclusive acquisition of the lock"). This is why Sun had to add the lock daemon. You can see some obscure breakages in code which assumes that this time-honored technique will always work, when that code is run on machines accessing the directory containing the lock file across an NFS mount. It's fun to explain this to people who have a couple of windows up, one a remote login to another machine which is also NFS-mounted on the local machine, and have "ls" of the same directory show two different things! -- Eric Black "Garbage in, Gospel out" UUCP: {sun!sunncal,hpda}!athertn!ericb Domainist: ericb@Atherton.COM
jc@minya.UUCP (John Chambers) (01/14/88)
In article <160@teak.athertn.Atherton.COM>, ericb@athertn.Atherton.COM (Eric Black) writes: > In article <448@minya.UUCP> jc@minya.UUCP (John Chambers) writes: > >In article <11156@brl-adm.ARPA>, GUTHERY%ASC%sdr.slb.com@RELAY.CS.NET (guthery%asc@sdr.slb.com) writes: > >> - systemwide semaphores > >I'll let the other wizards hack apart the rest of the items on > >the list; I'll just point out that every Unix I've ever seen has > >systemwide semaphores. They just aren't called that. Try: > > if ((l = creat("/usr/spool/locks/<somename>",0)) { > > ... > > [Critical section] > > ... > > unlink("/usr/spool/locks/<somename>"); > > } else { > > [Didn't get semaphore] > > } > >It worked on BRL Version 6, and it still works, even on SunOS. > ^^^^^ Hey, wait just a minute here; I didn't mention SunOS in my original posting; I said Sys5 and BSD. I wonder how I got misquoted? I have seen some things from people who said they tried it on their Foonix system and it works, so maybe some of the postings got jumbled together. Oh, well, it's not a real big deal, and lets us get into a session of Sun-bashing. (;-) > Careful in an NFS environment! It doesn't work (where "work" is defined > as "guarantees exclusive acquisition of the lock"). > > This is why Sun had to add the lock daemon. Yeah, in fact, Sun admits that NFS violates Unix file-system semantics in numerous ways. They say so in their manuals. Unfortunately, some of the things they don't implement are the things that you need in a distributed environment even more than on a single processor. Atomic reads, writes and file creation are among the most important. And they're not even all that hard to do right. You just have to avoid the temptation to speed things up by caching the same block in more than one place. > You can see some obscure breakages in code which assumes that > this time-honored technique will always work, when that code > is run on machines accessing the directory containing the lock > file across an NFS mount. The mistake is in porting code that depends on a Unix run-time environment to a non-Unix system (like SunOS). File-system stuff is a classical source of subtle problems, and when you go from Unix to SunOS, you have the worst possible case of a system which looks very familiar, but... (Of course, when Sun and AT&T produce their merged OS, it will be like Unix but will no longer support atomic I/O. :-) -- John Chambers <{adelie,ima,maynard,mit-eddie}!minya!{jc,root}> (617/484-6393)
mjy@sdti.UUCP (Michael J. Young) (01/15/88)
>In article <11156@brl-adm.ARPA> GUTHERY%ASC%sdr.slb.com@RELAY.CS.NET >>(guthery%asc@sdr.slb.com) writes: >>1) Here are some O/S goodies that stock OS/2 (now playing on my desk) >>has that stock Unix (also playing on my desk) doesn't have: >> - built-in light weight processes blended compatibly >> with heavy weight processes >> - runtime dynamic linking and demand loading >> - shared global memory segments >> - file locking by byte region >> - standard system calls from drivers >> - periodic signals You forgot some: - A toy file system - An inability to examine process status (a la ps(1)) - An inability to kill an errant process - A multitasking philosophy that permits the foreground process to preempt normal background processes (try running two compilations, one in the foreground screen group, one in a background one, and see how fast the background process completes!) - A command interface that is as nice as MSDOS (granted, CMD.EXE was designed to make people WANT the Presentation Manager) - EDLIN (yes folks, it's still there) Maybe if AT&T worked a little harder, they could create a new Unix standard that has some of these neat features (they could call it System "V--"). Note: :-) :-)! -- Mike Young - Software Development Technologies, Inc., Sudbury MA 01776 UUCP : {decvax,harvard,linus,mit-eddie}!necntc!necis!mrst!sdti!mjy Internet : mjy%sdti.uucp@harvard.harvard.edu Tel: +1 617 443 5779 "Are we having fun yet?" -- Zippy
boyd@basser.oz (Boyd Roberts) (01/15/88)
All this mindless blathering about OS/2 v UNIX sparked by Mr Pournelle. I'd hardly call him an authority on operating system design. From where I stand all I see is this guy with a doctorate in psych who writes sci fi. As for his ravings in Byte, I'd say they're primarily generated by the totally mindnumbing brain-damage of the micro world. Editor Y looses its mind over program X's file format. It may be entertaining, but its certainly not worth worrying about. ``I learnt my computer science from hacking out text on WordStar'' -- the mind boggles. As for OS/2. You want a software turd? Get Big Blue & Microsoft into gear and choose a micro with a brain-damaged architecture and THEN try for backwards (upwards) compatibility. I can't wait. Personally I'm for a new software foundation... NNU - NFS's not UNIX But for the moment, it's time for another Valium... Boyd Roberts boyd@basser.oz ``When the going gets wierd, the wierd turn pro...''
alex@umbc3.UMD.EDU (Alex S. Crain) (01/15/88)
>>>1) Here are some O/S goodies that stock OS/2 (now playing on my desk) ... >You forgot some: ... > - EDLIN (yes folks, it's still there) > >Maybe if AT&T worked a little harder, they could create a new Unix standard >that has some of these neat features (they could call it System "V--"). My AT&T 3b1 arrived without the development software, just the kernal and core utilites. While playing around with the window shell stuff, I ran into an editor in the menus. Selecting it produced ed, running in a window. -- :alex. alex@umbc3.umd.edu
cramer@clem.UUCP (01/16/88)
>I'd hardly call him [Pournelle] an authority on operating system design. >From where >I stand all I see is this guy with a doctorate in psych who writes sci fi. Well, it's a step up from having a doctorate in sociology ;-) >As for his ravings in Byte, I'd say they're primarily generated by the totally >mindnumbing brain-damage of the micro world. Editor Y looses its mind over >program X's file format. It may be entertaining, but its certainly not >worth worrying about. ``I learnt my computer science from hacking out >text on WordStar'' -- the mind boggles. Agreed. The fact that this cretin and his columns (which are filled with shameless plugs for his books and his son's enterprises, blather about the zillion computers he has at his house [how does he afford the electricity?], and dopey enthusiasms for whatever the greatest-computer-program-in-the- universe happens to be this month) are actually popular speaks volumes about the sorry state of computer-hobbydom. Sam Cramer {cbosgd,decwrl,hplabs,seismo,ucbvax}!sun!cramer cramer@sun.com
dave@stcns3.stc.oz (Dave Horsfall) (01/16/88)
In article <1176@basser.oz> boyd@basser.oz (PUT YOUR NAME HERE) writes: >All this mindless blathering about OS/2 v UNIX sparked by Mr Pournelle. >I'd hardly call him an authority on operating system design. From where >I stand all I see is this guy with a doctorate in psych who writes sci fi. Second that. He should stick to writing science fiction, where he shines. I know! Maybe I should try my hand at SF, and I'll be an authority too... -- Dave Horsfall (VK2KFU) ACS: dave@stcns3.stc.OZ.AU STC Pty Ltd ARPA: dave%stcns3.stc.OZ.AU@uunet.UU.NET 11th Floor, 5 Blue St UUCP: {enea,hplabs,mcvax,uunet,ukc}!\ North Sydney NSW 2060 AUSTRALIA munnari!stcns3.stc.OZ.AU!dave
lm@arizona.edu (Larry McVoy) (01/18/88)
In article <11156@brl-adm.ARPA> GUTHERY%ASC%sdr.slb.com@RELAY.CS.NET (guthery%asc@sdr.slb.com) writes: >1) Here are some O/S goodies that stock OS/2 (now playing on my desk) > .... >3) Kernel hacking is not research. A Unix wizard who doesn't know what >SIGOPS is loses five hit points. Do the wizards sense extinction? I'll bite on this one. And maybe (but probably not) lay it to rest. What we are and have been seeing for some years now is a merging of the good ideas from all operating systems into the new ones. It's not a question of "Unix vs. XYZ" rather "pipes vs. temp files". However, most good ideas seem to come out of Unix (not surprisingly, it has a very demanding user base) and it would do these young upstarts well to remember that. On the other hand, these ideas are usually a little rough... I suspect that in 10 years or so it won't make a whole lot of difference what machine you sit down to. While there will always be some (useful) oddballs, most OS's will have shells/tools that will make you feel at home right away. Many of those tools will be common across machine/OS boundaries. So what's the big deal, eh? -- --- Larry McVoy lm@arizona.edu or ...!{uwvax,sun}!arizona.edu!lm Use the force - read the source.
reggie@pdn.UUCP (George W. Leach) (01/21/88)
In article <207@stcns3.stc.oz> dave@stcns3.stc.oz (Dave Horsfall) writes: >In article <1176@basser.oz> boyd@basser.oz (PUT YOUR NAME HERE) writes: >>All this mindless blathering about OS/2 v UNIX sparked by Mr Pournelle. >>I'd hardly call him an authority on operating system design. From where >>I stand all I see is this guy with a doctorate in psych who writes sci fi. > >Second that. He should stick to writing science fiction, where he shines. >I know! Maybe I should try my hand at SF, and I'll be an authority too... Judging by the amount of discussion this guy and his views have generated in this newsgroup, one would have to assume that there is a good portion of people reading his stuff. If you don't like what he has to say or don't think he is qualified to be writing about it, then don't buy the rag he writes for. It is the most effective way to get your point across :-) -- George W. Leach Paradyne Corporation {gatech,rutgers,attmail}!codas!pdn!reggie Mail stop LF-207 Phone: (813) 530-2376 P.O. Box 2826 Largo, FL 34649-2826
cory@upba.UUCP (02/03/88)
>> 1) Here are some O/S goodies that stock OS/2 (now playing on my desk) >> has that stock Unix (also playing on my desk) doesn't have: >> [ Scott lists a bunch of things OS/2 has, that he thinks don't >> exist on Unix machines. ] > [ Many people show him his ignorance. ] >> 2) Microsoft/IBM marketing hype seems to bother some folks. I will >> wager a chocolate Dunkin' Donut that the number of dollars spent on hyping >> Unix divided by the number of years Unix has been out is GREATER >> than the number of dollars spent hyping OS/2 divided by the (fractional) >> number years its been out. ... Microsoft, IBM, and AT&T all launch big $$ ad campaigns when they want to promote things, especially new releases. Given the wager as Scott explains it, those months of little or no promo for new stuff by AT&T, compared to the fact promo bucks for OS/2 haven't started slowing down yet, clearly stack the deck against OS/2 and Scott. Looks like Scott has alot of Donuts to buy people... >> 3) Kernel hacking is not research. A Unix wizard who doesn't know what >> SIGOPS is loses five hit points. Do the wizards sense extinction? >> >> Cheers, Scott You have to wonder about someone who measure penalties in terms of "hit points". I don't think Wizards need worry about anything. -Cory ...!ihnp4!upba!cory (Cory Dekker @ United Phone Book Advertisers)| DISCLAIMER: Work: 1221 N St, Suite #800; Lincoln, NE 68508 {Ph: 402-476-2200}| My posting.. Home: 800 Foxcroft Ct, #188; Lincoln, NE 68510 {Ph: 402-483-7761}| MY opinions!
rroot@edm.UUCP (uucp) (02/03/88)
There are a lot of things that OS2 claims to be able to uniquely do, but here
are a couple of things that it CAN'T do (as far as I can tell) that it seems
UNIX CAN.
1) Run on a PS2/30
2) Run on a 386 machine IN 386 MODE.
3) work NOW on a '286 machine. (rather than just the command interpreter)
--
-------------
Stephen Samuel
{ihnp4,ubc-vision,seismo!mnetor,vax135}!alberta!edm!stevealex@umbc3.UMD.EDU (Alex S. Crain) (02/03/88)
>>> SIGOPS is loses five hit points. Do the wizards sense extinction? > You have to wonder about someone who measure penalties in terms > of "hit points". I don't think Wizards need worry about anything. Oh, I don't know, "hit points" seems as valid a way of relating unix penalties as anything. After all, rogue/larn/hack are unix games, the ports were an afterthought, and I've never met a wizard wo hasn't dedicated at least a few hours of their life to slaughtering screen characters with a broadsword. Tee Hee :-) -- :alex. nerwin!alex@umbc3.umd.edu alex@umbc3.umd.edu
ron@topaz.rutgers.edu (Ron Natalie) (02/09/88)
First off the PS2/30 is a piece of junk, it's an 8088. Running UNIX without protected mode is a lose. You just can't multitask well in that architecture. Second, there are at least two versions of UNIX out that will run on a 386 in 386 mode. The one I am most familiar with is the Intel (Interactive Systems) System V for the 80386 which seems to be a perfectly good UNIX as system V's go. Sun also has their new "roadrunner" product that run's Sun OS (a hybrid BSD UNIX). -Ron
tim@ism780c.UUCP (Tim Smith, Knowledgian) (02/10/88)
< There are a lot of things that OS2 claims to be able to uniquely do, but here < are a couple of things that it CAN'T do (as far as I can tell) that it seems < UNIX CAN. < 1) Run on a PS2/30 < 2) Run on a 386 machine IN 386 MODE. < 3) work NOW on a '286 machine. (rather than just the command interpreter) Yeah, well here are some things that The Exec ( that was Mattel's OS in the Intellivision ) could do nearly a decade ago that Unix still can't do: 1) Run on a GI 1600 2) Work with only a few hundred bytes of RAM ( and no disk ) 3) Asynchronous IO at the application level -- Tim Smith, Knowledgian tim@ism780c.isc.com "Who needs sex, drugs, and rock n' roll when you've got Missile Command?" -- Anon.
jc@minya.UUCP (John Chambers) (08/01/88)
In article <11156@brl-adm.ARPA>, GUTHERY%ASC%sdr.slb.com@RELAY.CS.NET (guthery%asc@sdr.slb.com) writes: > 1) Here are some O/S goodies that stock OS/2 (now playing on my desk) > has that stock Unix (also playing on my desk) doesn't have: > ... > - systemwide semaphores > ... I'll let the other wizards hack apart the rest of the items on the list; I'll just point out that every Unix I've ever seen has systemwide semaphores. They just aren't called that. Try: if ((l = creat("/usr/spool/locks/<somename>",0)) { ... [Critical section] ... unlink("/usr/spool/locks/<somename>"); } else { [Didn't get semaphore] } It worked on BRL Version 6, and it still works, even on SunOS. > 3) Kernel hacking is not research. A Unix wizard who doesn't know what > SIGOPS is loses five hit points. Do the wizards sense extinction? Nah, as long as vendors keep delivering systems with such pieces of cruft as sendmail, vi/emacs, NFS, and the threatened OSI stuff, our income is secure. It's even better than OS/360 was: It's even fun to work with. (:-) Kernel hacking CAN qualify as research. I've spent many happy hours trying to come up with reliable ways of automating the diagnosis of problems; some of the most effective ways involve getting into the kernel and inserting some spies that write to audit files. I mean, there's lots of systematic, understood methodology for dealing with computational and communication processing; there is very little for dealing with when things don't work right. About all we have are ad-hoc, after-the-fact tools. There's a real need for intelligent, systematic tools for the job, especially now that we are getting into the mass confusion called "networking". [BTW, the signal(2) in my manual here doesn't seem to include SIGOPS; what version of Unix are you running on? :-] -- John Chambers <{adelie,ima,maynard,mit-eddie}!minya!{jc,root}> (617/484-6393)