From: Digestifier To: Linux-Misc@senator-bedfellow.mit.edu Reply-To: Linux-Misc@senator-bedfellow.mit.edu Date: Mon, 23 Aug 93 05:13:08 EDT Subject: Linux-Misc Digest #43 Linux-Misc Digest #43, Volume #1 Mon, 23 Aug 93 05:13:08 EDT Contents: Re: A question about the SLS sysinstall script's copyright (Mark Lord) Re: ASCII to PostScript converter (Telly Mavroidis) Re: NT versus Linux BRIAN LEARY KILLED KENNEDY!!! (James Hightower) Re: NT versus Linux, the updated and expanded comparison chart, version 2 (Kevin Andrew Buhr) Re: NT versus Linux (Bernd Meyer) Re: Why would I want LINUX? (Drew Eckhardt) Re: NT versus Linux (Drew Eckhardt) ---------------------------------------------------------------------------- From: mlord@bnr.ca (Mark Lord) Subject: Re: A question about the SLS sysinstall script's copyright Date: Mon, 23 Aug 93 03:14:53 GMT In article <251q8u$bfl@news.u.washington.edu> kenney@stein.u.washington.edu writes: > >In article <930819232631.23132@lambada>, >Steve McMahon wrote: >>In article <1993Aug20.015821.15839@cc.gatech.edu> byron@cc.gatech.edu (Byron A Jeff) writes: >>>In article <930819205847.23148@lambada>, >>>Steve McMahon wrote: >>> >>>>Peter cannot do that. It's against the GPL. >>> >>>Yes he can. He wrote the scripts. They are his. >>> >> >>No he cannot. These scripts were made specifically for the interface >>and installation of a GPLed product (linux), hence they fall under the >>GPL. If Peter made them for OS/2 or NT it would be another matter, but >>not here. >> > >Huh???? Does this mean if I package up some GPL code with tar (not >GNU tar) that all of a sudden tar falls under the GPL. Sounds like a >lot of nonsense to me. tar would be exempt, but the SLS scripts would not be: "For an executable work, complete source code means all the source code for all modules it contains, plus any associated interface definition files, plus THE SCRIPTS USED TO CONTROL COMPILATION AND INSTALLATION of the executable. However, as a special exception, the source code distributed need not include anything that is normally distributed (in either source or binary form) with the major components (..) of the operating system on which the executable runs, unless that component itself accompanies the executable." tar would be exempt, the SLS install scripts would be covered. -- mlord@bnr.ca Mark Lord BNR Ottawa,Canada 613-763-7482 ------------------------------ From: mavroidi@acf2.nyu.edu (Telly Mavroidis) Subject: Re: ASCII to PostScript converter Date: 23 Aug 1993 03:44:30 GMT Ghostscript comes with gslp, that's supposed to be like enscript. -- **************************************************************************** Telly Mavroidis mavroidi@acf2.nyu.edu ^-leave the last s off for savings. **************************************************************************** ------------------------------ Crossposted-To: comp.os.ms-windows.advocacy From: jamesh@netcom.com (James Hightower) Subject: Re: NT versus Linux BRIAN LEARY KILLED KENNEDY!!! Date: Mon, 23 Aug 1993 03:54:09 GMT brileary@world.std.com (Brian Leary) writes: Lots of stuff Could someone 'splain to me why Mr. Leary's posts come to us by some distinctly different rout than any other messages from world.std.com? Why his message ID's have a format different from the one that all other messages from world.std.com have? Why fingering brileary@world.std.com returns "%No matches for brileary"? I smell a conspiracy! A coverup! Where's Oliver Stone when you need him? Bet seriously, I havn't really been following this thread, and I'm sure you've beaten the subject of B.L.'s forgeries to death by now, but I think the fact that someone has gone to this much trouble to discredet Linux shows that it's existince is making an impact. When the WABI and Wine stuff are done, MS may find itself in the position of having a competitor that it can't sue. Sue who? Linus? Demand all the profits? Brian Leary, in some wierd way, reminds me of David Sternlight. JJH ------------------------------ Crossposted-To: comp.os.ms-windows.advocacy From: buhr@umanitoba.ca (Kevin Andrew Buhr) Subject: Re: NT versus Linux, the updated and expanded comparison chart, version 2 Date: Mon, 23 Aug 1993 05:18:48 GMT In article kevin@frobozz.sccsi.com (Kevin Brown) writes: | | Feature NT Linux | ------- -- ----- . . . | kernel+necessary | extensions to run ~500k+ 200k-500k This is why NT runs in a minimum 1 Meg configuration, right? 500K for the system with 500K to spare for apps, eh? ;) The NT "kernel + necessary extensions to run" is presumably somewhere in the order of 7 or 8 Megs, or it would run comfortably in less than 12 Megs. Kevin Buhr ------------------------------ Crossposted-To: comp.os.ms-windows.advocacy From: root@umibox.hanse.de (Bernd Meyer) Subject: Re: NT versus Linux Date: Mon, 23 Aug 1993 01:58:27 GMT brileary@world.std.com (Brian Leary) writes: >A few neutral observations in this subject: >Observation #1: >Making fun and degrading remarks about Windows users is fashionable in >the linux groups. Linux users have the moronic idea that they must be >somehow mentally and morally superior to the rest of the world. Snide >remarks about how stupid and gullible Winodws users are are all but >uncommon in the linux groups. I believe you are somehow mistaken here - it's more about the OS/GUI in itself than about its users that I see such remarks. Granted, I only read about 10% of c.o.l... >Observation #2: >The above is mainly due to the strong SHEEP MENTALITY running in the >linux community. Few if any linux users have their own mind, After >all, why bother? It's easier to just echo what a master sheep has >said. This is obvious in their "rebuttals" to the check list that was >posted here. Each one just echoes the other -- all of them asserted >that linux has QIC-80 support, just because the first one said so, >even though linux HAS NO QIC-80 SUPPORT, nor is likely to have it any >time soon. Oh my god - not again. If you don't believe linux has QIC-80 support, then call +49 40 7654276, log in as guest and do "findsoft ftape". It's there. It's working. People use it to do backups. What else do you need to believe it? So though I agree that many linux users are somehow, well, over-confident about their OS, most if not all responses I have seen so far were right on technical aspects (this is far more than I can say about the NT supporters side). >But that's not all, there is also the LYNCH MOB MENTALITY. After >hearing about how dum and gullible we are for two years, as soon as an >objective comparison that shows that linux is not the ultimate os is >posted here, the linux lynch mob mentality takes over, and we're >burried by an article after another of vulgar flames. Even though the >comparison was not posted to their group. Well, how would you react if I posted an article claiming that NT won't run in under 64MB, won't run DOS or Windows apps more complicated than "hello world", uses 16 bit 286 protected mode, doesn't support the majority of SCSI cards and costs $15000 to a linux group, and somebody else crossposted a reply to this group? >Of course, there is also the FANATIC MENTALITY, but that one is >obvious. Right. One of the greatest dangers linux is in - that fanatic users try to thrust it upon people it is really not ready for yet (not the other way round :-). >Observation #3: >For a long time they kidded us about how NT is vaporware. The general >consensus among them is that NT will never see the light of day, and >our replies that it is coming out "soon" were laughed at. But look at >what they are saying right now: Loadable device drivers... "soon", There. Called modules. Call the above mentioned site, do a "findsoft module". It's there. It's working (the qic-80 support uses it :-). What else do you want? > Running ODS apps... >"emulator is coming along nicely", Well, I said this before - show me how you run lemmings 2 under NT, and I'll admit that you have an emulation that is at least equal with linux'. Until then, be careful what you declare vaporware. BTW, this time do "findsoft dosemu" :-) > Running Winodws apps... "WABI is almost here" Yeah - the development has gone on for exactly two months now - and I can download a new experimental set every week. This isn't what I call vaporware! Something is vaporware when you speak about it on CeBit '90, show a version "Beta, close to shipping" on CeBit '91, show another version "Gamma, we just have to tweak that last bugs out" at CeBit '92, and introduce it to the market with lots of PR on CeBit '93, and several months later really start shipping..... Been there, heard it! >Observation #4: >Linux is inferior to DOS! How's that? Well, DOS has loadable device >drivers, linux doesn't. see above. "findsoft modules" >Your printer and modem use the same IRQ? No >problem in Windows, just don't use them at the same time. In linux, >you have to use the modem a little, stop, recompile the kernel with >the printer enabled and the modem disabled, use the printer a little, >stop, recompile, use the modem, stop, ...etc. Two drivers that use the >same IRQ cannot live in the same kernel, even if none of them is used! This isn't true (OK, I haven't tested with parallel ports, but I have used a 3434 four serial setup during the last 5 weeks. It works! BTW, why should I have interrupts enabled for my parallel port if not using PLIP? If using PLIP, the kernel is expecting IP-packets every microsecond... >Observation #5: >Not only that linux doesn't support drag an drop, its users have no >concept of what drag and drop is! Most of them think it's just cut and >paste, as can be seen clearly in the "rebuttals". OK, I was among those that answered along this line. To tell you the truth, at least I have a rather clear concept of WHAT IT IS. Only, I have no idea as to what it is good FOR.... This all comes down to filemanagers in the end (surprise surprise... at least on of the FMs for linux supports D&D :-), and I strongly doubt that they improve efficiency. >Observation #6: >Linux users have no idea what multimedia support is. Multimedia >support is NOT that the os is able to talk to a sound card! DOS can >do that. It is that the os has inherent layering facilities to do >multimedia. To run a full motion picture clip or play a MIDI song via >the sequencer in Windows, you just call a single API function (e.g. >MciSendCommand). OK, this is nice - BTW, how much memory is wasted when the driver for that gimmick is dynamically loaded? The unix approach is more along the lines "if we have a tool for it, we don't waste any energy on reimplementing it. And if we can do it in userspace, we don't bloat the kernel with it". You want a full motion video? OK, write something like "this software requires that mpeg_player is available on the system. It is available from xxx.yyy.zzz and compiles out of the box", then simply spawn a shell executing it. Same for sound (ever heard of .mod-files? How do you play them under NT? Oops :-). What is needed at the kernel level (sound support, CD support) is there, the user programs are there ("findsoft sound", "findsoft mpeg"), what else do we need? >Observation #7: >QIC-80 support is not in linux, nor is it ever likey to be. The >general consensus among linux developers is that QIC-80 is simply >unsuitable to multitasking environments, a claim that's obviously >disproved by NT's support for QIC-80. Oops, didn't I read this several pages above? You seem in desparate need to find a long list! BTW, if you think about a certain "Jesus Monroy Jr", he is a pain in the *ss, seemingly unable to produce working code. Also, he isn't a linux user, let alone developer.... >Observation #8: >Q: What does linux do when it runs low on memmory? A: IT FALTERS! The >only os I have seen that locks up when it runs low in memory. Every >linux user must constantly watch for memory useage, if linux needs a >few more kilobytes than are available, it doesn't complain, it just >crawls to a halt before dying a miserable death, taking down the whole >system with it. Was true until, well, I think about a month ago. Well, nearly true. It first started throwing away the program code, then started to swap. If you invested 16MB for swap, it was REALLY hard to have the machine crash because of low memory. But this is history, linux now handles this situation more gracefully (plus - I wonder what NT does? Ooh, you mean if I malloc 16MB under NT, it really takes 16MB to fulfill my wishes?!?). >Need I say more? There is still lots to talk about: the flaky file >system, the feeble networking, ..etc. ..etc. You obviously haven't used a current version of linux for at least half a year. I haven't used NT ever. I don't think I'm in the know to flame NT. Why do you think you are for linux? >Linux users, get a clue! Your os is shit. And remember, we didn't >start this. Well - you did. One of you. He posted a really ridiculous list comparing the two. most I have seen from the linux side has been technically correct, though the tone sometimes wasn't very nice. Calling what runs on my computer "shit" isn't that nice, either! And what your side said about linux was mostly plain wrong.... Bernie -- We both know that the earth is round | Bernd Meyer, EE-student So we can't see the way before us to its end | "Nobody is a failure who has We walk on this way, hand in hand, | friends" (from: isn't it a And I hope you are still with me behind the horizon| wonderful life?" ------------------------------ Crossposted-To: comp.os.386bsd.misc From: drew@kinglear.cs.colorado.edu (Drew Eckhardt) Subject: Re: Why would I want LINUX? Date: Mon, 23 Aug 1993 07:47:49 GMT In article <258gu6INNlef@bonnie.tcd-dresden.de> j@bonnie.tcd-dresden.de (J Wunsch) writes: >In <24vd7h$frk@horus.mch.sni.de> Martin.Kraemer@mch.sni.de (Martin Kraemer) writes: >[for Linux] >>Plus there is much more support for "cheap" hardware and for two-or- >>more-OS's-on-one-harddisk. > >Yep, the Linux support for two-or-more-OS's ain't the best. When my system comes up, my master boot record (winiboot) asks me which partition I want to boot. If I don't choose something (with one keypress) within the timeout period, Linux is automatically booted.. If I press some key for a different OS, it gets booted instead. How can multi-OS operation be simpler than that? While this isn't the default installation (with LILO), winiboot is available as part of the shoelace package, available in source and binary form. >i hate Linux, but i'm still curious why they decided to have this silly >off-the-standard booting scheme (LILO). LILO is a bootblock replacement for the bootblock within a partition, there's nothing non standard about that. As far as the LILO bootblock being different from BSD's - it was developed under a different set of design constraints. Size is more limited than with BSD because we don't have things sub partitioned with an area for the disklabel / bootblock set aside. This limits the amount of functionality. We can't work like the BSD bootblock and prompt to boot any file, since users must be able to boot of more than one type of FS, and we don't have room to support the proliferation of filesystems in the bootblock. So, we needed a minimal bootstrap that could locate a block mapping for various kernel immages and load them as directed. >With *BSD using a normal dozz >boot scheme (load MBR, and then load the active partition's boot sector), That's how Linux + LILO and the master boot record of your choice. >along with one of the fancy boot managers (i'm using os-bs), it works >like a charm. You could also boot those silly unices requiring their >own partition being marked active. The same thing applies to Linux. >(Another problem of Linux is, they occupy a full dozz partition for swap instead of sub-partitioning their primary one.) Linux supports MS-DOS extended partitions, if you choose to use them you can have multiple Linux / DOS / etc. partitions in one physical partition. >Btw., i've been working with my old cheep PC from older dozz times. As are many starving college student and unemployed hacker types :-) >The really disadvantage of BSD is it's lack of shared libs, thus consu- >ming much more disk space. But the original shared libs from Linux didn't >convince me either: i saw it at a friend, he quickly felt that his Linux >got binary-incompatible to itself. (Since the binaries had to match >exactly the shared libs.) The key word there is "original". The original implementation of shared libraries under Linux was deficient in this regard, since they were statically linked. This got you very low disk space usage, low memory usage, etc, but none of the benefits of dynamic linking and ease of fixing library bugs in all your binaries. In other words, it was still better than no shared libraries. After that, Linux got the so-called "jump table" libraries, where the jump table was at a fixed location but the routines could move. This let us upgrade shared libraries without recompiling anything, making it easy to fix library bugs. The currenty Linux shared library implementation is dynamic. > >Last not least: take out all the `unnecessary' things from the BSD >kernel (IP, various file systems, SCSI, Ethernet, SLIP etc. etc.), >you'll get a (IMHO much useless) very tiny kernel:-) The same thing applies to all unices. Any good system administrator will trim out everythign he/she doesn't need to maximize the amount of free memory and consequently the size of the incore page set. -- Boycott USL/Novell for their absurd anti-BSDI lawsuit. | Condemn Colorado for Amendment Two. | Drew Eckhardt Use Linux, the fast, flexible, and free 386 unix | drew@cs.Colorado.EDU Will administer Unix for food | ------------------------------ Crossposted-To: comp.os.ms-windows.advocacy From: drew@kinglear.cs.colorado.edu (Drew Eckhardt) Subject: Re: NT versus Linux Date: Mon, 23 Aug 1993 08:06:18 GMT In article kevin@frobozz.sccsi.com (Kevin Brown) writes: >In article <1993Aug21.013647.21842@colorado.edu> drew@romeo.cs.colorado.edu (Drew Eckhardt) writes: >>In article <1993Aug20.184709.15303@mnemosyne.cs.du.edu> zevans@nyx.cs.du.edu (Zack Evans) writes: >>>Are you absolutely sure that you want to do this? [ny] no, but I am going to >>>anyway :) >>> >>>In article <930820201354.23255@world>, >>>Brian Leary wrote: >>> >>>>Q: What does linux do when it runs low on memmory? A: IT FALTERS! The >>>>only os I have seen that locks up when it runs low in memory. Every >>>>linux user must constantly watch for memory useage, if linux needs a >>>>few more kilobytes than are available, it doesn't complain, it just >>>>crawls to a halt before dying a miserable death, taking down the whole >>>>system with it. >> >>This is a false statement. > >From looking at the code, it was true of pl10. Here's the relevant code: pl10, two publicly distributed revisions and over half a dozen alphas behind the current code. If you're allowed to critisize Linux based on older versions, can I ream NT based on older software, say Windows 3.0? [ deleted ] >Sbrk() is a library call. It may do some kind of additional checking, but >after perusing the code, I don't see any happening in libc-4.4.1. So what >you see in sys_brk() is what you get. Check it out in your kernel sources >to see what it's doing in the version you're running. ( after various rlimit checks, code from .99.12, it was originally added to one of the pre-.99.11 alpha releases and was generally available in .99.11 ) freepages = buffermem >> 12; freepages += nr_free_pages; freepages += nr_swap_pages; freepages -= (high_memory - 0x100000) >> 16; freepages -= (newbrk-oldbrk) >> 12; if (freepages < 0) return current->brk; So, we figure free memory as the sum of truly free pages, size of the buffer cache, and how much sawp we have left. We deduct some based on memory size for padding, figure out how much would be left if this brk() succeeded, and if that value is negative, fail. >>>I have fond memories of Windows 3.1 doing this... and FYI it doesn't do this >>>quite so badly at patchlevel 11 and higher. >> >>Of course, recent Linux kernels have fixed this behavior. > >Must be pl11 or above... It was available in alpha patches against .99.10, .99.11 was the first publically available distribution that handles things correctly. -- Boycott USL/Novell for their absurd anti-BSDI lawsuit. | Condemn Colorado for Amendment Two. | Drew Eckhardt Use Linux, the fast, flexible, and free 386 unix | drew@cs.Colorado.EDU Will administer Unix for food | ------------------------------ ** FOR YOUR REFERENCE ** The service address, to which questions about the list itself and requests to be added to or deleted from it should be directed, is: Internet: Linux-Misc-Request@NEWS-DIGESTS.MIT.EDU You can send mail to the entire list (and comp.os.linux.misc) via: Internet: Linux-Misc@NEWS-DIGESTS.MIT.EDU Linux may be obtained via one of these FTP sites: nic.funet.fi pub/OS/Linux tsx-11.mit.edu pub/linux sunsite.unc.edu pub/Linux End of Linux-Misc Digest ******************************