From: Digestifier To: Linux-Misc@senator-bedfellow.mit.edu Reply-To: Linux-Misc@senator-bedfellow.mit.edu Date: Thu, 6 Oct 94 12:13:22 EDT Subject: Linux-Misc Digest #888 Linux-Misc Digest #888, Volume #2 Thu, 6 Oct 94 12:13:22 EDT Contents: Re: Yggdrasil Linux Plug and Play CD ver1.1 ? (Paul Bash) Help for NCR 53C810 SCSI disk & Video ATI-68800 chip set (Pradeep Chetal) Diamond Stealth 64 PCI drivers (Allen S. Harris) Re: xvnews (Hans de Graaff) Re: SW Technologies (Jonathan I. Kamens) Re: Word (Text) processors for Linux? (J.J. Paijmans) ---------------------------------------------------------------------------- From: bash@tware.com (Paul Bash) Subject: Re: Yggdrasil Linux Plug and Play CD ver1.1 ? Date: Wed, 5 Oct 1994 16:48:24 GMT In article , Jeff Kesselman wrote: >> Funny, Slackware, OS/2 and DOS (all installed on the same disk) don't >> see any problems. I have no other option except skipping the disk >> partition phase of the install. > >I'm honestly not sure on this one, but I seem to recalll there being a >patch bandied about thats needed for this particular controlelr to run >two drives?? (Someone with more experience with Adpatecs feel free to jump >in.) Linux has no problem with this host adapter. I've been running Slackware on it for close to a year now. I've run SVR4, OS/2 and DOS on it for years before that. SCSI host adapters _always_ support more than two devices (up to 7 in fact, 8 if you count the adapter itself). Its part of the SCSI specification. The OS using the adapter, of course, must have support for the additional drives offered before they can be used. The INT 13 BIOS on the Adaptec provides the basic support DOS needs for the first two drives. Beyond that _DOS_ needs additional driver support not supplied with the OS. Perhaps that is what you are thinking. I believe, nay I'm positive, that this has no bearing on the problem. Linux has all the additional support necessary right out of the box (providing you are using the SCSI kernel) and Linux has _excellent_ support for the 1542B. I suspect that this has more to do with the master boot record or the partition table not being to the liking of the fdisk used in Yggdrasil. It is probably checking some form of signature in the MBR that doesn't match up to what it expects. The OS/2 boot manager is active on this drive and that might be confusing things. The OS/2 boot manager, though, is nothing new to Linux users. There are notes in various README's that tell how to have Boot Manager and LILO co-exist on the same disk. You install boot manager first, then you don't allow LILO to take over the MBR during the Linux install. This is dirt simple and has worked for a _long_ time. The fdisk used in Slackware sees nothing wrong with the disk. Perhaps it is different from the one used by Yggdrasil. I can't imagine why but I guess its a possibility. The Yggdrasil fdisk is obviously less well tested than Slackware's if it is in fact the culprit. And no, this is _not_ another example of how Yggdrasil isn't meant for someone such as myself. It is just this ideal Yggdrasil audience you talk about that would be likely to install it on the same disk with the OS/2 boot manager... just to try it out while continuing to do their normal work on OS/2 (or any other OS on the same disk). > >By the way, the hoops you have to jump through to 'get rid of that damn >cd-rom' are: > umount /dev/system_cd >And it all goes away. > Oh, don't start getting smug now, Jeff. We were doing so well and now you have to go and make and _effort_ to piss me off. You don't have a clue of what you are talking about here, but you continue to act like you do, inserting foot into mouth in the process. Sad. (yes, I'm being excessively condescending, even an asshole, but Jeff just keeps pushing :-) First of all, that should be umount /system_cd You umount a _directory_ and mount a _device_. There is no /dev/system_cd. There _is_ a /dev/cdrom0 in Yggdrasil (and its link, /dev/cdrom). This is just standard UNIX symantics. Ok, now that we've got the command straight, have you actually tried this? It doesn't just "all goes away". Jesus, haven't you read this thread yet? More than one person has complained about how, when you do a complete install from the CD-ROM, you _still_ have symbolic links all over the file system that point at directories under /system_cd. You can't just remove the CD-ROM without removing all the links that point to it. Those links, most likely, have to be replaced by the CD-ROM files they are pointing to else you don't have a clean install. You might get away with running without the CD-ROM for awhile, but that's like saying you can walk down the middle of the freeway at 5pm and not get hit by a car. Sure you can... for a randomly short while. I ask you again, have you actually tried this? For more than a couple of days? While exercising all those neat packages you installed? Since you've commented elsewhere that you are quite happily running from the CD-ROM without installing everything on your PC, I doubt it. Here's an extreme, but highly plausible, illustration of the problem: what if I want to access the QRZ Ham Radio CD-ROM but can't because the system needs the Yggdrasil CD-ROM mounted to, say, run the C compiler executing in the background? Well, I guess I just can't, right? Sounds like a classic Catch-22. Sure, I could buy a second CD-ROM drive just for this application, but then my $35 "Plug and Play" has become a $335 "Plug and PAY" system. Yggdrasil just got a whole lot more expensive. This issue of extraneous symbolic links is just one of several that started this whole thread long before I got involved. If you don't know this, you did more than miss a turn a while back, you've been asleep at the wheel. If you don't understand the ramifications of this, yet continue to throw smug comments around as if you do, you're just making yourself look silly. Like I said, you need to do some homework. Here's your first lesson: get a 300MB+ disk and, using the control panel, install the full set of software from the CD-ROM onto your disk. Now issue the following commands: cd / umount /system_cd find / -type l -ls | grep system_cd On my system, I installed the following subset from the control panel (after using the command line install to setup usrbin) since I didn't have the 300MB to work with: UUCP and USENET GNU development tools GNU C and C++ Compilers GNU Emacs ghostscript Kernel Sources Elm and Pine mailers Man page sources Portable bitmap system XF user interface builder Other X Windows Programs Xview I then applied the Fall 94 errata fixes. When I issue the above commands, I get quite a list of links that still point at the /system_cd directory. After editing that list to remove those things that I _think_ are accounted for by those packages I _didn't_ install (whose links shouldn't be there, but should be harmless provided I don't attempt to run those packages) I have the following links that look suspicious: /usr/lib/g++-include -> /system_cd/usr/lib/g++-include /usr/lib/nslookup.help -> /system_cd/usr/lib/nslookup.help /usr/bin/adduser -> /system_cd/usr/bin/adduser /usr/bin/man- -> /system_cd/usr/bin/man- /usr/X386 -> /system_cd/usr/X386 /usr/account_template -> /system_cd/usr/account_template /usr/g++-include -> /system_cd/usr/g++-include /usr/i486-linux -> /system_cd/usr/i486-linux /usr/man/host.1 -> /system_cd/usr/man/host.1 /usr/man/man9 -> /system_cd/usr/man/man9 /.bash_history -> /system_cd/.bash_history /INSTALL -> /system_cd/INSTALL /bootflpy.3in -> /system_cd/bootflpy.3in /bootflpy.5in -> /system_cd/bootflpy.5in /bootflpy.phl -> /system_cd/bootflpy.phl /fips11.zip -> /system_cd/fips11.zip /manual -> /system_cd/manual /ramdisk -> /system_cd/ramdisk /readme.txt -> /system_cd/readme.txt /rr_moved -> /system_cd/rr_moved /setup -> /system_cd/setup Out of this list, a few items catch my eye as particularly annoying: /usr/lib/g++-include -> /system_cd/usr/lib/g++-include /usr/lib/nslookup.help -> /system_cd/usr/lib/nslookup.help /usr/bin/adduser -> /system_cd/usr/bin/adduser /usr/bin/man- -> /system_cd/usr/bin/man- /usr/X386 -> /system_cd/usr/X386 /usr/account_template -> /system_cd/usr/account_template /usr/g++-include -> /system_cd/usr/g++-include /usr/i486-linux -> /system_cd/usr/i486-linux /usr/man/host.1 -> /system_cd/usr/man/host.1 /usr/man/man9 -> /system_cd/usr/man/man9 In particular, look at /usr/lib/g++-include. Try this command cd /usr/lib/g++-include BINGO! It's not found, is it? That doesn't sound right, Jeff. I requested that the C and C++ compilers packages be installed on my hard disk. Well, there goes my stable C++ development environment. Is a light coming on somewhere, Jeff? If the links are there, then you don't have a CD-ROM-less install. You have an accident waiting to happen. It doesn't matter if you are a power user or a beginner, the situation is just as messy. From his view, random programs are going to blow up and the target Yggdrasil user you defined earlier isn't going to have a clue as to what is happening. Ok, yes, I could just go through and manually fix all of this. That's not the point. The point is/was that this is sloppy and poorly executed. Particularly when the Yggdrasil manual indicates throughout that you have the option of mounting the CD-ROM or not (if you at least install the /usr/bin package). With a CD-ROM, you can't afford to miss these details because you don't have the option of just patching the distribution when you find a problem. Instead, the user just has to deal with it again and again every time she installs the system. I've personally been through the 2 page Fall 94 errata 5 or 6 times (each time I've had to re-install the CD-ROM to try and get a clean system). I hope I never see it again. If Yggdrasil intends a CD-ROM-less installation option, they have to test it, dammit! It doesn't look like they did and that's sloppy. I found these problems in 5 minutes by looking for them, why didn't they? Slackware doesn't have these kind of problems (although it has had _some_ problems). It is _much_ more skillfully executed. And, it has been this clean for every release I've seen (since 1.1 and it is now at 2.1, I believe). FMPE (From My Personal Experience), the great majority of users will find it satisfying much longer than they would the Yggdrasil CD-ROM. Thus, we come full circle to my original comments that you jumped on several posts ago thereby fueling this fire. Yggdrasil is "cute" but it isn't for serious use. Jan experienced some of the same problems I did and asked if a better Linux CD exists. Without going into 20,000 words (like I've had to with you) I told him what I thought. Do you see what I mean now, Jeff? I hope so, but I kind of doubt it. >> >>I wasted $35 on the Yggdrasil Fall 94 CD-ROM that I will never use. > >I have a feeling that, if you are that unhappy, and you bought it >directly from yygdrasil theyw ill probobly refund your money. I did not buy this software directly from Yggdrasil. Instead, I bought it from an exhibitor at a Hamfest. Third party, cash deal. I don't expect Yggdrasil to refund my money on this. If they would, great, but I don't expect it. And, it wouldn't change my opinion of Yggdrasil anyway. I have no beef with Yggdrasil as a whole, I just said the Fall 94 CD-ROM had major problems. I'm sure they're a great group otherwise and I'm sure they'll eventually get it together. They obviously have the talent. The execution is what's lacking. >> >>(If you aren't interested in Jeff and I bitching at each other, please press >>"n" now) >> >>Back to Jeff's comments: >>------------------------ >> [stuff deleted] > >This is kind of childish debate tactics (here comes the return-flame..) >Paul, and beneath your otherwise very intellegent comments above. You just couldn't resist getting that last little dig in, eh? That's beneath you, Jeff ;-) -- Paul Bash Techware Design bash@tware.com Boulder, CO U.S.A. "The Net interprets censorship as damage and routes around it" -- John Gilmore ------------------------------ Crossposted-To: comp.os.linux From: chetal@gedny.ml.com (Pradeep Chetal) Subject: Help for NCR 53C810 SCSI disk & Video ATI-68800 chip set Date: Wed, 5 Oct 1994 20:57:06 GMT Hi, I just installed Linux on a DELL machine with slackware from sunsite. I had to pick up a modified ncr roootdisk for SCSII NCR 53C810 disk, which was there called ncr.gz and I picked it up and it does WORK!! BUT when I create the boot disk from setup, the kernel there is NOT capable of NCR SCSI. How can I update the system kernel & boot disk kernel to be same as the root disk 'ncr' kernel. In addition, I have an ATI 68800-AX chip set with NEX 4FGe. Can anyone send the Xconfig file for it. I keep on getting errors about the clock speed. Thanks, /Pradeep -- -- Pradeep Chetal Internet: chetal@gedny.ml.com -- -- Pradeep Chetal Internet: chetal@gedny.ml.com ------------------------------ From: asharr@cs.wm.edu (Allen S. Harris) Subject: Diamond Stealth 64 PCI drivers Date: Thu, 6 Oct 1994 07:55:32 GMT I purchased a new computer a couple of months ago including a Diamond Stealth 64 PCI video card. I have since decided that linux is a very good thing, and would like to put it on my machine. Problem: xfree doesn't support the Diamond Stealth 64. I don't blame them, but I would still like to run x in extended video modes (1024x768 would be nice). Does anyone know what options are available to me? ie are there specs available for the Diamond Stealth 64 so that I might (gulp) write a driver for it. I know that Diamond's non-disclosure policy kind of hampers that possibility, but I am hoping that there are other people out there who would like to/ are running linux with the Stealth 64. Any help would be _greatly_ appreciated. Thanks, Scott Harris -- email at: asharr@cs.wm.edu ------------------------------ From: graaff@dutiws.twi.tudelft.nl (Hans de Graaff) Subject: Re: xvnews Date: Thu, 6 Oct 1994 08:20:45 GMT In article <36ubu7$28f@scitsc25.wlv.ac.uk>, J.Tench wrote: >Dear all > Does any one know where I can get hold of a copy of the source for xvnews. ftp.twi.tudelft.nl:/pub/news/xvnews-2.2.1.tar.gz ftp.uu.net:/networking/news/readers/xvnews/xvnews-2.2.1.tar.gz Hans -- Hans de Graaff J.J.deGraaff@TWI.TUDelft.NL Delft University of Technology Department of Information Systems ======================================================================= click me ------------------------------ From: jik@cam.ov.com (Jonathan I. Kamens) Subject: Re: SW Technologies Date: 6 Oct 1994 15:26:04 GMT In article <37028n$e0p@hk.super.net>, shciosea@hk.super.net (Mr John Shaller) writes: |> Something from Jonathan saying that "...I'll keep reposting the message |> until they admit their fault or out of business..." make me feel awkward. |> I don't there is such big deal to justify forcing someone until they die That is not why I am posting my review periodically. I am posting my review periodically because there are constantly new people deciding to buy a system and looking for a vendor to use, and those people often only start reading the relevant newsgroups when they make the decision to buy, and I want to warn them to avoid a vendor which I believe is not competent to sell and support the systems they claim to be able to sell and support. I no longer believe that I have much of a chance to get back any of the money I've asked SWT to refund. I am no longer posting my review to get back money. I am posting my review to prevent other people from being damaged as I was by SWT. |> Personally, I think the requirements from Jonathan is just too much. |> Everyone in the Linux world know that Linux is provided "AS IS", you try |> it at your own risk. A number of people have made this point to me, in postings and in E-mail. However, I think the people who make this point are missing the major thrust of my complaints about SWT. If the problems I'd had with the machine they'd sold me had been only with Linux, I wouldn't be complaining at all. I'm aware that Linux is "AS IS" software and that most Linux installations have rough edges; in fact, that's one of the reasons I want to run Linux -- to play with smoothing those rough edges (and, in fact, in the time I had the machine from SWT, I made a number of fixes to the Linux kernel and submitted them to Linus). The problems which prompted me to return the machine, and which prompt me to believe that SWT is not competent to sell Pentium PCs (I can't speak directly to their competence to sell other PCs, but the anecdotal evidence I've seen seems to suggest that they're slightly better at 386's and 486's than at Pentium machines), were all related to HARDWARE and SERVICE. The hardware I got from SWT was faulty when I got it and was faulty when I returned it three months later. SWT's attempts to fix the faults were feeble and incompetent. The hardware would have been faulty whether I was running Linux, MS-DOS, Windows or OS/2. I mention in my review problems with the software that SWT installed only to make the point that they were part of a larger series of problems. In and of themselves, they would not have prompted me to return the machine. |> I can see |> that Marvin has been trying hard to help by shipping replacements and |> suggesting importments... Nothing in the world is perfect... Marvin tried, but he tried in an incompetent manner. As I believe I said in my review, the fact that SWT tried hard to fix serious problems with the machine does not change the fact that the machine should not have had those problems when it was shipped to me. Would you consider it acceptable if you bought a new car and spent three months driving back to the dealership almost daily to get things fixed? There are new cars which don't cost much more than I paid for the computer from SWT, so I believe the analogy is completely reasonable. |> A final word, Jonathan, PLEASE DON"T periodically reposting your LONG |> statement ( and the finely tunned correction/amandment... are we in the |> California court?). Just try to think there are how many news server |> around the world and your REPEATED posting may consume a few GB of disk |> space :-).... I've already explained above why I repost my review. If you don't want to see it, then put its Subject line in your KILL file. Somehow, I don't think that my posting of a 26Kb article to a few newsgroups every three months is going to swamp the net. I'll leave to the reader the proof that the resources taken up by such posting are negligible (read "in the noise"). -- Jonathan Kamens | OpenVision Technologies, Inc. | jik@cam.ov.com ------------------------------ From: paai@kub.nl (J.J. Paijmans) Crossposted-To: comp.unix.questions Subject: Re: Word (Text) processors for Linux? Date: 6 Oct 1994 09:06:56 GMT In article <1994Oct5.174859.18757@midway.uchicago.edu> goer@midway.uchicago.edu writes: ... > >Those are all ISO 8859-1 languages. You can get those without even >trying. I'm talking about things like Japanese, Arabic, Greek, and >so on. While multilingual in one sense, apps that do the languages >you're talking about are basically just using a single (Latin) script. > >Guys, the race is on to capture growing markets in China, India, and >perhaps Russia and Islamic countries, and Unix is way behind the Mac >(WorldScript) and NT (Unicode); probably behind NeXTStep, too, though >I don't know what they've been doing lately.... > Richard: I don't understand. OK, you can get Hebrew or Arabic under Windows (I even saw a wordprocessor for old-egyptian hieroglyphs demonstrated), but surely they are just translations of graphics for existing characters? You can't even write hebrew in the right direction (i.e. from right to left) when you select the font. If you want to do that, you have to start from scratch and break out that assembly language manual; or at least the toolbox with graphic functions and there is not much in MS-Windows that you can use right away. The hieroglyphic wordprocessor was done this way. So if the waiting just is for somebody to draw a new font, I see no intrinsic superiority of MS-Windows (or MacIntosh) over Unix. And if you want to mess with the basic left-right orientation, the situation in X Windows is not worse than in MS-Windows. But I am not an expert in this sort of things, so please enlighten me. Paai. ------------------------------ ** 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 ******************************