From: Digestifier To: Linux-Development@senator-bedfellow.mit.edu Reply-To: Linux-Development@senator-bedfellow.mit.edu Date: Mon, 6 Dec 93 04:13:06 EST Subject: Linux-Development Digest #292 Linux-Development Digest #292, Volume #1 Mon, 6 Dec 93 04:13:06 EST Contents: Re: SQL for University Ingres (Roy Hann) Re: THE ULTIMATE FAQ ANSWER (Marc Andreessen) lockd for Linux? (Michael C. Loewen) Mainstream development (Re: scroll-back tty's) (Frank Lofaro) Re: scroll-back tty's (Stefan Nehlsen) Re: xdm & keyboard problem (Stefan Nehlsen) Re: Creeping featuritis (post --rant --flame) (Elizabeth Haley) Re: scroll-back tty's (Matthew J. Ryan) Re: Xconfig (Barry Lynam) A Linux Users Manual (DAVIN GEORGE) HELP: ext2fs-0.4a lost directory ! (Gerd Rausch) Re: xdm & keyboard problem (Hendrik G. Seliger) tcp header checks? (Gene Choi) Re: Linux and 1.72M floppies (3 1/2" HD) (Frank Damgaard) Re: Neuronal networks (Rene Jager) ---------------------------------------------------------------------------- Crossposted-To: comp.databases,comp.os.linux.misc From: rhh@tachy.uah.ualberta.ca (Roy Hann) Subject: Re: SQL for University Ingres Date: Sun, 5 Dec 1993 21:08:16 GMT In article <8h0UNYC00VR54Gw3Qe@andrew.cmu.edu> you write: : zmbenhal@netcom.com (Zeyd M. Ben-Halim) writes: : > I seriously doubt that anyone is going to add SQL to university ingres, : > at least without being paid for it. Most of the people familiar with : > QUEL and SQL think QUEL is better and have no incentive to add SQL. : : Buzzt! You lose. : : Ingres is our main production base in my department, and there isn't : anyone here who would rather use QUEL than SQL. My guess would be : that the choice of whether to use QUEL or SQL would be determined by : personal preferences, portability, and availability. I refuse to believe that anyone who knows QUEL would rather use SQL if they had the free choice--ie they are not constrained by external considerations like recruiting and maintainability. Are you trying to tell us you know QUEL but actually LIKE SQL? Frankly that defies belief. I can't believe it. I don't believe it. On the other hand, if you are just saying SQL is "standard" and widely available and as a matter of policy you use it for those reasons, well: ho-hum; you and me both. But that does not contradict Zeyd. QUEL is the better language--it's not popular, but it IS better--and no one would rip it out of University Ingres to put SQL in, because (I conjecture) no one using University Ingres is going to be concerned about maintainability etc. and the virtues of QUEL are manifold and manifest. So, Buzzt yourself, and thanks for playing. PS: This could shape up to be an amusing Holy War. If the flames start to fly it might be fun to move this thread to c.d.ingres ======================================================================== Roy Hann Senior Analyst, Information Systems rhh@tachy.uah.ualberta.ca University of Alberta Hospitals (MIME-capable mail agent) WMC 2C2.21, 8440-112th Street, Edmonton, Alberta Tel: (403)492-4367 T6G 0N4 FAX: (403)492-3090 Canada ======================================================================== ------------------------------ From: marca@ncsa.uiuc.edu (Marc Andreessen) Crossposted-To: comp.infosystems.www,comp.windows.x,comp.windows.x.i386unix,comp.windows.x.motif,gnu.misc.discuss,comp.sources.d Subject: Re: THE ULTIMATE FAQ ANSWER Date: 5 Dec 93 15:56:46 In article pcg@aber.ac.uk (Piercarlo Grandi) writes: Ahhhh, but this is the crucial aspect of the problem: as a rule OSF want people to pay for copies of their _runtime_ system too, while everybody else is content with being paid just for their _development_ system. Are you completely clueless? From Darrell Crow's recent posting: ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 2) The Software Developer statically binds in the Motif "Runtime Copy" (as defined in the license supplement). Applications are shipped with Motif "Runtime Copies". [...] B. Developer is a Full Distribution Rights Licensee of OSF/Motif or has a University Site License. License Reqs.: The application distributed by the licensee (developer) may be installed on any system, whether or not the system already has a Motif license. Royalty Reqs.: No royalties will be due OSF in this instance. ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Marc -- Marc Andreessen Software Development Group National Center for Supercomputing Applications marca@ncsa.uiuc.edu (MIME welcomed here) ------------------------------ From: mloewen@cpumagic.scol.pa.us (Michael C. Loewen) Subject: lockd for Linux? Date: Sun, 5 Dec 1993 23:33:44 GMT I've seen this question at least twice within the last couple of months, but never an answer: has anyone ported rpc.lockd over to Linux? Or developed a workable substitute? Sun's PC-NFS package can mount UNIX filesystems on PCs, with optional file locking support, if the remote server has rpc.lockd running. For now, it doesn't appear that you can mount Linux directories on a PC, IF any of the PC applications require file locking. -- Mike Loewen The Centre Programmers Unit BBS mloewen@cpumagic.scol.pa.us (814) 353-0566 V.32bis/HST ------------------------------ Crossposted-To: comp.os.linux.misc From: ftlofaro@unlv.edu (Frank Lofaro) Subject: Mainstream development (Re: scroll-back tty's) Date: Sun, 5 Dec 93 23:39:10 GMT In article <126972@hydra.gatech.EDU> gt8134b@prism.gatech.EDU (Robert Sanders) writes: > >A more complete implementation could be put in the kernel, but I'm ^^^ >not sure the benefits would be worth the bloat. I can usually get ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ >the same effects using pagers when I need scrollback. Other >approaches are to run a shell under GNU Emacs or screen. > Follow-ups to comp.os.linux.misc. Ouch!!!!!!!!! That means that if someone decides to implement it, there will people rallying against it?! I sure hope not! Maybe make something like that configurable. I don't like to see development discouraged. I myself and a friend have quite a few things we would like to see in Linux. Things like raw devices, async I/O, etc, etc. Things that would make Linux a more complete operating system. We (and other contributers), would be willing to implement stuff ourselves, but certainly not if it isn't going to be in the main kernel distribution. Hacking something into a kernel to get a feature I need is acceptable. (I am willing to implement features I need/want. I am not going to just sit back and whine and do nothing useful. I can understand developers frustration at those type of users.) But: Hacking something into every kernel that comes out to get a feature I need is _NOT_ acceptable! In other words, please give us small-time developers a chance to contribute. It will make us all much happier and Linux much better. If it works, let it go into the kernel, unless there is a _very_ compelling reason to do otherwise. If it is questionable, just put a config option to enable it. This is _NOT_ a flame, merely a suggestion. Follow-ups to comp.os.linux.misc. ------------------------------ From: stefan@nehlsen.toppoint.de (Stefan Nehlsen) Subject: Re: scroll-back tty's Date: Mon, 6 Dec 1993 01:11:54 GMT boswell@chem.canterbury.ac.nz (Ross Boswell) writes: >Laszlo Herczeg (las@whome.uucp) wrote: >: What would it take to implement scroll-back on tty output? >: That is, the user could scroll back on the screen output by hitting >: a certain key, such as PgUp. >No problem at all -- just hit Shift/Page-Up and watch it happen! Nobody told me, I think 99% of all Linux users don't know this! bye Stefan -- Stefan Nehlsen nelli@toppoint.de ------------------------------ From: stefan@nehlsen.toppoint.de (Stefan Nehlsen) Subject: Re: xdm & keyboard problem Date: Mon, 6 Dec 1993 02:27:19 GMT boy@atm.cm.nctu.edu.tw () writes: >Hello, there, > I have installed SLS 1.03 and recently upgrade the XFREE to 2.0 > It seems just fine. And recently I have tried to put the xdm to /etc/rc > but found that when I logged in, under the xterm, the 'numbersign' and 'at' > cannot be typed in xterm. All I have got is that SHIFT-3 acts like > 'backspace' and SHIFT-2 acts like 'CTRL-U'. I use the xmodmap -pk to see > if the key is being redefinied or not. But the results shows just normal. >% xmodmap -pk >There are 4 KeySyms per KeyCode; KeyCodes range from 8 to 134. > KeyCode Keysym (Keysym) ... > Value Value (Name) ... > 8 > 9 0xff1b (Escape) > 10 0x0031 (1) 0x0021 (exclam) > 11 0x0032 (2) 0x0040 (at) > 12 0x0033 (3) 0x0023 (numbersign) > 13 0x0034 (4) 0x0024 (dollar) > 14 0x0035 (5) 0x0025 (percent) > 15 0x0036 (6) 0x005e (asciicircum) > 16 0x0037 (7) 0x0026 (ampersand) > 17 0x0038 (8) 0x002a (asterisk) > 18 0x0039 (9) 0x0028 (parenleft) > 19 0x0030 (0) 0x0029 (parenright) > 20 0x002d (minus) 0x005f (underscore) > 21 0x003d (equal) 0x002b (plus) > 22 0xff08 (BackSpace) > 23 0xff09 (Tab) >. It happen to me too. I'm running SLS1.03 with XFree86 V2.0. Starting X11 by xdm makes xterms behave strange. `nn' will not run !!! Values given by `stty -a': virtuell term xterm (startx) xterm (xdm) speed 9600 9600 0 intr ^C ^C ^? erase ^? ^? # kill ^U ^U @ min 1 1 0 <---- causes the nn-problem i suppose cs8 cs8 cs5 <---- this don't look good hupcl + + - cread + + - brkint + + - ignpar + + - ixany + + - `xdm' will not run `/etc/profile' and '~/.profile', but this should happen on every login. `xdm' set's $PATH to a strange default value. That's all i found out till now (2:20AM). Tomorrow I will go on ( so don't flame if i forgot something) Is there somebody who knows more? let me hear! Maybe we should discuss this in an other group. bye Stefan -- Stefan Nehlsen nelli@toppoint.de ------------------------------ Crossposted-To: gnu.misc.discuss Subject: Re: Creeping featuritis (post --rant --flame) From: haley@scws14.harvard.edu (Elizabeth Haley) Date: 6 Dec 93 01:40:33 GMT NNTP-Posting-Host: scws14.harvard.edu Xref: husc.harvard.edu gnu.misc.discuss:1127 comp.os.linux.development:3080 grante@hydro.rosemount.com (Grant Edwards) writes: [large amount of stuff deleted to avoid repeating it *again* ;-) ] >When discussing the problem commonly known as "gnubloat" it is wise >to remember the famous indictment: > "RMS thinks that Unix is just another program you can run under emacs." Hehehehehehe. Actually, I have thought about it, and it would be relatively possible to add a few device drivers to emacs, and maybe some system calls, and you could just have your bootblocks point straight to the emacs binary. Might even make it a little faster. It can already be used as a shell... >Thus the now-famous gnu attitude "let's throw in 3 or 4 kitchen sinks, >we've got virtual memory! Oh, and put in a lisp interpreter -- if >we've forgotten anything, we'll add it on later..." ;) >Back in the PDP-11 V7 days, the limit of 64K text and 64K data required >a certain amount of elegance finesse that seems to be lacking lately. Agreed. On the other hand, I can do without the days of memories too small to handle a medium sized data-set. My goal in memory size has always been to be able to have the entire Manhattan phonebook in memory, as well as the OS and my favorite programs... We're almost there... -- Hacksaw ------------------------------ From: ryanm4@hall101.its.rpi.edu (Matthew J. Ryan) Subject: Re: scroll-back tty's Date: 6 Dec 1993 03:10:37 GMT In article <1993Dec6.011154.477@nehlsen.toppoint.de>, Stefan Nehlsen wrote: >boswell@chem.canterbury.ac.nz (Ross Boswell) writes: > >>Laszlo Herczeg (las@whome.uucp) wrote: >>: What would it take to implement scroll-back on tty output? >>: That is, the user could scroll back on the screen output by hitting >>: a certain key, such as PgUp. > >>No problem at all -- just hit Shift/Page-Up and watch it happen! > >Nobody told me, I think 99% of all Linux users don't know this! > Hmm...Is this little lifesaver documented anywhere (other than the source :) ) And are there any other interesting keyboard controls like it? I took a quick look through my documentation collection, didn't see it, though I might have overlooked it.... - Matt -- "Still looking for a funky quote..." Matthew Ryan ryanm4@rpi.edu ------------------------------ From: lynam@maize.qut.edu.au (Barry Lynam) Subject: Re: Xconfig Reply-To: B.Lynam@qut.edu.au Date: Mon, 6 Dec 93 04:50:59 GMT The only thing "wrong" with modeDB.txt, IMHO is that the monitor timing should have the sync rates in it so that new modes can be generated for different resolutions. Barry +--------------------------------------------------------------------- ---+ | Barry Lynam EMail: B.Lynam@qut.edu.au | | Communications Software Support Fax: +61 7 864 1343 | | Computing Services Postal: GPO Box 2434 | | Queensland University of Technology Brisbane 4001 | | Brisbane AUSTRALIA AUSTRALIA | +--------------------------------------------------------------------- ---+ ------------------------------ From: davin.george@welcom.gen.nz (DAVIN GEORGE) Subject: A Linux Users Manual Date: Mon, 6 Dec 1993 00:09:00 GMT Gidday Fokes. As a MsDos user who had problems himself with installing Linux I figured it wouldn't be a bad idea if somebody wrote basically a Installation and Command Reference Book for new Linux users. I've started work on it already and was wondering what people would want in a book like this. I mean we've no doubt all gone out and brought Unix books and found that some of the commands just don't apply and spent the next day trying to find something remotely similar. Okay - so you're going now after working it out but wouldn't you rather have had a quick reference book of some type. If you're like myself you probably didn't even know about the FAQS. Okay myself I seem a bit denser that the average user but I didn't have access to the Internet at the time. I wouldn't want anybody else going through what I did to get started with Linux, Would you? Well I've made a start on a reference book but I've a few questions to ask that you can answer that would help make if easier for people in the future to go to Linux from Dos or Os/2. Your help with this would be greatly appreciated. 1. What commands do you think would be of most benefit to the new user and how involved would you want the commands? 2. What type of format would you want the book arranged in and a few reasons as to why? What sort of chapters would you find useful. ie Installing Linux, Linux Commands etc. 3. Would you want if taylored to suit a specific brand of Linux ie Yggdrasil, Sls etc or just in general with references to each specific brand of Linux. 4. How about a Dos Command/Linux Command equivalent section? Would you find this useful if you knew the Dos command and could just look up the equivalent Linux command. 5. How about a list of available software for Linux with info like where to get it and what it does. Great advertising for you software authors out there. If you want your software included in this Email me at Davin.George@f370.n771.z3.fidonet.org along with a description of the software, version number, any specific requirements ie needs SLS 99.13.25 and the filesize and where it can be obtained from. 6. Any specific topics you'ld like covered in detail ie Installing and Running UUCP etc. Okay so this is going to take a while to do so I allowing for it not being finished for about 18 Months. If you a know a specific topic or subject backwards and would like a bit of recognition for it Email me. Any help would be greatly appreciated. Catch Ye Later Davin George ///////////////////////////////////////////////////////////////////// // // // Internet: Davin.George@f370.n771.z3.fidonet.org // // NetMail: 3:771/370 // // // ///////////////////////////////////////////////////////////////////// WinQwk 2.0 a#0 ------------------------------ From: eedger@teamos41.ericsson.se (Gerd Rausch) Subject: HELP: ext2fs-0.4a lost directory ! Date: Sun, 5 Dec 1993 23:13:33 GMT Some days ago I upgraded to linux-0.99pl14 and ext2fs-0.4a After some reboots and one e2fsck (maxmimal mount number reached ...) I noticed, that one directory has been completely lost. Notice, that I've shut down the system correctly every time, and it never crashed. There was only one real checking from e2fsck in startup phase because of the maximal mount number, but it only reported me about some error in the bitmap (sorry i can't recall that one exactly, but it was something about a free entry marked as used). Where is my directory ? Fortunately it's been something reconstructable. If the directory is not there anymore, at least e2fsck should be able to find it and put it into /lost+found. Any pointers ? (Note: After this happened, I reconstructed the whole system from tape, which I've backup'ed 2 days before) -- Don't let the ".se" fool you ____/| Gerd Rausch \ o.O| =(_)= voice: +49-2407-575-353 U email: eedger@aachen.ericsson.se ------------------------------ From: hank@Blimp.automat.uni-essen.de (Hendrik G. Seliger) Subject: Re: xdm & keyboard problem Date: Mon, 6 Dec 1993 07:57:04 GMT Reply-To: hank@Blimp.automat.uni-essen.de boy@atm.cm.nctu.edu.tw wrote: : Hello, there, : I have installed SLS 1.03 and recently upgrade the XFREE to 2.0 : It seems just fine. And recently I have tried to put the xdm to /etc/rc : but found that when I logged in, under the xterm, the 'numbersign' and 'at' : cannot be typed in xterm. All I have got is that SHIFT-3 acts like : 'backspace' and SHIFT-2 acts like 'CTRL-U'. I use the xmodmap -pk to see : if the key is being redefinied or not. But the results shows just normal. It seems that if called from xdm xterm has some tty-settings which are not normal. I had the same problems. It's easily solved by puting the following line into your /usr/X386/lib/X11/app-defaults/XTerm: *ttyModes: intr ^c quit ^\ kill ^u erase ^? Hope this helps, Hank -- ====================================================================== Hendrik G. Seliger Universitaet Essen hank@automat.uni-essen.de Schuetzenbahn 70 Tel.: +49-201-183-2898 45117 Essen, Germany ====================================================================== "Handling interrupts is simple." (G. Pajari) "Interrupts are an unpleasant fact of life." (A. Tanenbaum) ------------------------------ From: genie@scam.Berkeley.EDU (Gene Choi) Subject: tcp header checks? Date: 6 Dec 1993 08:40:54 GMT Hi there, I'd like to report a small problem I've been having with (C)SLIP using linux 0.99pl11-pl14. This might be a problem with the way tcp packet headers are assembled (although I don't have a clue really since I haven't read the source or know even the faintest details behind it). Anyway here's the problem: CSLIP is working perfectly when I turn on the error correction capability of my modem - a 14.4k Zoom. (error correction being either v.42 or MNP5). However, when I start up (C)SLIP without error correction, after about 10 minutes of plug and chug, the computer freezes and/or reboots. Again, this does NOT happen when I turn on error correction. My best guess is that the problem arose because of some random line noise creeping into the phone line. Anyway I reported to my school sysadmin this problem I was having - he reported that it was probably they way linux did tcp packets, and that random line noise was doing something nasty to the OS. He also reported that SunOS way back when had this exact same problem - without error correction, the kernel went berzerk with bad packets via line noise. My problem is this - shouldn't random noise be ignored and packets resent without freezing/rebooting the kernel? I wish I could pinpoint what the problem was but I can't since I have no xp with net code in any way. Now you ask, why use CSLIP with no error correction? Because response time is WAY WAY WAY better WITHOUT error correction (reduces my ping times by 100-120 ms && takes me under the 200 ms mark for 64 byte pings). You can actually get work done without error correction (in my opinion). Correction added too much latency times making the text jumpy and bothersome. My setup now: Linux pl14 386 33 12 megs ram 16550AFN com2 (modem). but problem existed from pl11. -Gene genie@xcf.berkeley.edu ------------------------------ From: frank@diku.dk (Frank Damgaard) Subject: Re: Linux and 1.72M floppies (3 1/2" HD) Date: Mon, 6 Dec 1993 08:29:39 GMT Using 1.72M floppy disks (and other formats). Under Linux 0.95 and 0.96 I patched the appropriate drivers, and boot boot code. This is not necessary under Linux 0.99p13. If a 1.72M 3.5" floppy is wanted , ( 2 sides, 82 tracks/side , 21 sectors/track = 3444 sectors / 1722 Kb): 1) format the disk under DOS (with fdformat) (It should be possible to format floppy disks under Linux with fdformat, but I don't have the correct parameters.) 2) # from the shell or in /etc/rc or /etc/profile ... : setfdprm -p /dev/fd0 3444 21 2 82 0 0x0C 0x00 0x0CF 0x6F # mounting floppy on /drv_a with msdos file system. mount /dev/fd0 /drv_a -t msdos It is also possible to create and mount a minix filesystem (14 or 30 char. filenames). I have tried creating an Ext2 filesystem on a floppy, but the overhead is to big on smaal devices, so I wouldn't recommend this. --- Frank Damgaard University of Copenhagen E-Mail frank@diku.dk Dept. of Computer Science ------------------------------ Subject: Re: Neuronal networks From: rene_j@tuderz.et.tudelft.nl (Rene Jager) Date: 6 Dec 93 09:41:06 +0100 Reply-To: R.Jager@ET.TUDelft.NL In article <13837@dirac.physics.purdue.edu>, bcr@bohr.physics.purdue.edu (Bill C. Riemers) writes: |> In article <1993Dec4.161139.22431@arthax.satlink.net> floyd@arthax.satlink.net (Christian Pablo Tagtachian) writes: |> >Hello, I am looking for a developement package for neuronal networks, |> >does anyone know about the existance of such thing for Linux? |> >I also got SmallTalk with its interface for X, but it doesn't allow you to |> >develope X11 applications, is there any add-on for it to support such |> >thing? |> >Thank you very much. |> >Christian |> > |> > Buenos Aires, Argentina. |> |> I'm not sure why you posted the same message twice, but anyways would |> something like JETNET work? It should compile with f2c and seems about |> one of the most intelligent neural network code I've used. If you like |> I can look up the e-mail address so you request the latest version. |> |> Bill |> |> try SNNS: Stuttgart Neural Network Simulator; works under Linux/X11 ------------------------------ ** 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-Development-Request@NEWS-DIGESTS.MIT.EDU You can send mail to the entire list (and comp.os.linux.development) via: Internet: Linux-Development@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-Development Digest ******************************