From: Digestifier To: Linux-Activists@senator-bedfellow.mit.edu Reply-To: Linux-Activists@senator-bedfellow.mit.edu Date: Mon, 16 Aug 93 09:13:08 EDT Subject: Linux-Activists Digest #124 Linux-Activists Digest #124, Volume #6 Mon, 16 Aug 93 09:13:08 EDT Contents: Re: A Word Processor for Linux (Malcolm Beattie) networking with pl12 (Daniel Jimenez) [Q] Can't do networking with SLS 1.03 (Ronald Aarts) Re: ctrl-alt-del in .99pl9-1 (Miquel van Smoorenburg) Re: Is this becoming comp.linux.advocacy? (Larry D Snyder) Problem with EXT2 (Frank Westheider) casts (was SCO cc is broken?) (Chris Torek) SLS 1.03 and XS3 (solved)!! (Alex Poon) Re: [Q] Diffs from SysV specs? (Dale Gass) [Q] Can't do networking (Ronald Aarts) Re: Modem and noise problems - Please help (Floyd Davidson) multiusercard and driver searched !!! (Thurecht) Re: Is this becoming comp.linux.advocacy? (Bill Vermillion) ---------------------------------------------------------------------------- From: mbeattie@black.ox.ac.uk (Malcolm Beattie) Subject: Re: A Word Processor for Linux Date: Mon, 16 Aug 1993 10:27:50 GMT In article <1993Aug13.222046.29390@kf8nh.wariat.org> bsa@kf8nh.wariat.org (Brandon S. Allbery) writes: >In article <24glm2$m40@nic.umass.edu> verdant@twain.ucs.umass.edu writes: >>Is there any reason to write the proposed Word Processor as a stand-alone >>program. I say this because, from what little I have seen of emacs, it seems >>to be a very flexible basic editor with a lot of potential for expansion. > >We had this flamewar last time :-( Emacs is just too big and has too much >baggage. I even suggested starting with a cut-down emacs and was rejected. >I think it's safe to say that this is a no-go. > >++Brandon >-- >Brandon S. Allbery kf8nh@kf8nh.ampr.org bsa@kf8nh.wariat.org Here's a concrete suggestion for a word processor. In brief: * Uses Tk/Tcl * Object oriented. Separate parts (tables, multi-column, whatever you fancy) can be written completely separately by anyone and dropped into a central kernel. * Allows export as TeX, PostScript, raw ascii or whatever else each object can cope with. * I volunteer to think about the central kernel and the associated `easy' bits: fonts, justification, headings and so on. Here are the details: fairly simple but fairly powerful, I think, and allowing the `distributed programmers' approach which is the hallmark of a lot of Linux hacking. Tk/Tcl: I came across them a few weeks ago and am extremely impressed. Easy to get started with; rapid prototyping (everything can be a script and the performance is fine even with scripts bound to mouse motion events); can recode anything in C (possibly C++ too) if wanted. Have a central kernel which is basically a geometry managing widget and deals with a few global menus. It copes with vboxes and hboxes just as TeX does and manages widget children (perhaps with glue in between). It can start off as a hacked version of Tk's packer or canvas. Each `part' of the document is an autonomous instance of a widget. Once the cursor is in a particular widget it `does its thing.' Standard text and text justification can be handled by Tk's text widget, appropriately configured. These text widgets already allow tagging of arbitrary ranges of characters with tag names and then configuring per tag name with font, relief, colour and so on. The interface between such a widget and the (word processor) kernel requires things like the following: Widget to kernel: How big am I? How big can I get? The user is searching for ..... and I haven't got it. Please pass on the search to the next widget up/down/left/right. The user is trying to move the cursor out of me. Please pass on the focus/cursor to the appropriate widget. Please add an entry ... to a global menu and call me back when it's used. Kernel to widget: Resize to such-and-such size. What formats of data can you export? Export your data as ascii/PostScript/TeX... User requested ... from a global menu that you registered. It's this interface that needs specifying and extending. Once that's in place, anyone can come along with a Tcl script file and say: "Put this file in directory /usr/local/lib/tkwp" and the next time you fire up your word processor you'll have such-and-such new facility. As I've said, I'll do some thinking about this and see what sort of facilities this kernel needs and how closely it ought to follow TeX's ideas of vboxes and hboxes (glue and all?) Since I am fairly good with TeX and know how excellent its text layout algorithms are, I definitely want the ability to export any reasonable document straight through TeX. A separate `object' (or configured text widget) to allow raw TeX commands to be included in the document is trivial to add. Comments welcome. --Malcolm -- Malcolm Beattie | I'm not a kernel hacker Oxford University Computing Services | I'm a kernel hacker's mate 13 Banbury Road, Oxford, OX2 6NN (U.K.) | And I'm only hacking kernels Tel: +44 865 273232 Fax: +44 865 273275 | 'Cos the kernel hacker's late ------------------------------ Crossposted-To: comp.os.linux.help From: djimenez@ringer.cs.utsa.edu (Daniel Jimenez) Subject: networking with pl12 Date: Sun, 15 Aug 1993 22:45:14 GMT I am having problems networking with patch level 12. My setup worked with pl11 and libc 4.4.1, now I can't get it to work with pl12 and libc 4.4.1 or 4.4.2. The ethernet route commands come back with `network is unreachable.' If I tell Linux to use netmask 255.255.0.0 (which is wrong for my network), then I can get things to work for a while, then they break after a few minutes (e.g., telnet sessions get rudely disconnected). Here is my routing rc: ---begin Dan's rc--- #! /bin/sh # # rc.inet1 This shell script boots up the base INET system. # # Version: @(#)/etc/rc.d/rc.inet1 1.01 05/27/93 # # Author: Fred N. van Kempen, # MYNAME=`/bin/hostname` NETMASK="255.255.255.0" IPADDR="129.115.10.31" NETWORK="129.115.10.0" # utsa2 GATEWAY="129.115.3.253" # Attach the loopback device. /etc/ifconfig lo 127.0.0.1 netmask 255.0.0.0 /etc/route add 127.0.0.0 /etc/route add 127.0.0.1 # Set up the Ethernet connection(s). /etc/ifconfig eth0 ${IPADDR} netmask ${NETMASK} /etc/route add ${NETWORK} /etc/route add ${IPADDR} /etc/route add default gw ${GATEWAY} metric 1 # All done. ---end Dan's rc--- I know pl12 is supposed to be more `strict' about routine, but I don't know what that is supposed to mean. It looks to me like I am doing things right, but I must not be since things aren't working. Any suggestions? I'd like to move permanently to pl12, but I need the networking stuff. Thanks! -- Daniel Jimenez djimenez@ringer.cs.utsa.edu "I've so much music in my head" -- Maurice Ravel, shortly before his death. " " -- John Cage ------------------------------ From: ronald@lowt.phys.tue.nl (Ronald Aarts) Crossposted-To: comp.os.linux.help Subject: [Q] Can't do networking with SLS 1.03 Date: 16 Aug 1993 11:18:04 GMT I still have problems getting my network going with the latest SLS 1.03 release (disk a1.3 dated Aug 12). In doinstall I select option 5 for NFS-install and I specify my IP-number, network number, etc. My NFS-host is on the same subnet. It looks as if the network configuration is OK (no errors or warnings), but the NFS-mount doesn't work: ! mount clntudp_create: RPC: Port mapper failure - RPC: Unable to send My ifconfig and route look like: ! lo IP ADDR 131.155.110.55 BCAST 131.155.255.255 NETMASK 255.255.255.0 ! MTU 2000 METRIC 0 POINT-TO-POINT ADDR 0.0.0.0 ! FLAGS: 0x0049 ( UP LOOPBACK RUNNING ) ! ! eth0 IP ADDR 131.155.110.55 BCAST 131.155.255.255 NETMASK 255.255.255.0 ! MTU 1500 METRIC 0 POINT-TO-POINT ADDR 0.0.0.0 ! FLAGS: 0x0043 ( UP BROADCAST RUNNING ) ! ! Kernel routing table ! Destination net/address Gateway address Flags RefCnt Use Iface ! network * UH 0 43 eth0 ! softland * UH 0 0 lo I don't think the latter is correct !!! After that I installed Linux from my hard disk, but I still couldn't get the network going. ifconfig and route give similar output as above and NFS-mounts give the same error. I can ping to localhost, but I can't ping to any other host. From the other side, if I try to ping to my PC from e.g. an RS/6000 I don't get a reply, but now a host with IP-address 0.0.192.31 shows up in the arp-table of the RS ???? I also tried the hints of Darcy Boese in comp.os.linux/53085: /etc/ifconfig eth0 ${IPADDR} netmask ${NETMASK} broadcast ${BROADCAST} /etc/route add default gw ${GATEWAY} metric 1 /etc/route add ${IPADDR} /etc/ifconfig lo 127.0.0.1 /etc/route add 127.0.0.1 but now the first two route commands complain about "network unreachable". I give up. Any hints ??????? -- Ronald Aarts Eindhoven University of Technology | Internet: ronald@lowt.phys.tue.nl Physics Department, P.O. Box 513, | Phone: +.31.40.474213 5600 MB Eindhoven, The Netherlands | Fax: +.31.40.453587 ------------------------------ From: miquels@caution.cistron.nl.mugnet.org (Miquel van Smoorenburg) Subject: Re: ctrl-alt-del in .99pl9-1 Date: Sun, 15 Aug 93 13:20:01 In article wtm1@crux2.cit.cornell.edu (Tark McMains) writes: >I want to make ctrl-alt-del run halt instead of reboot. Where is this >keystoke defined to executer reboot? Is it possible to change? >I'm running .99pl9-1 > > Look in /etc/inittab, and read the man page for shutdown. (And inittab) Mike. -- Miquel van Smoorenburg, Ibmio.com: cannot open CONFIG.SYS: file handle broke off. ------------------------------ Crossposted-To: comp.unix.misc,comp.unix.sys5.r4 From: larry@gator.oau.org (Larry D Snyder) Subject: Re: Is this becoming comp.linux.advocacy? Date: Mon, 16 Aug 1993 12:38:46 GMT evan@telly.on.ca (Evan Leibovitch) writes: >SCO WordPerfect ran out of the box on Esix, with the exception of a >glitch in how it dealt with the console. I'm curious, are you running this same WP (what release) under Unixware/Solaris? -- Larry Snyder Internet: larry@gator.oau.org Orlando, Florida UUCP: ..!uunet!tarpit!gator!larry ------------------------------ From: higgins@sisley.uni-paderborn.de (Frank Westheider) Subject: Problem with EXT2 Date: 16 Aug 1993 13:47:02 +0200 Hi Folks ! I us EXT-2-FS on my Linux-Box and i'm happy with it. But yesterday, i connected my Box via Ethernet (WD8003) and the pl9-Kernel to another box (NE200, pl9), and ... after working about 1 hour with nfs etc. and then checking my filesystem there were lots of errors (bit-set etc.). Is this a known bug ? Is this bug gone in pl12/11 ? CIAO Higgins ------------------------------ From: torek@horse.ee.lbl.gov (Chris Torek) Crossposted-To: comp.lang.c Subject: casts (was SCO cc is broken?) Date: 16 Aug 1993 12:00:06 GMT In article <1993Aug12.035402.7680@ksmith.com> keith@ksmith.com (Keith Smith) writes: >This is a kinda goofy construct here and I don't remember the precedence >of type casting as defined by ansi, ... The semantics of a cast are actually quite simple. The result of a cast is the same as if the cast had been an assignment to an unnamed temporary variable with the type given by the cast. Thus: char ch; unsigned char uch; int i; unsigned int ui; i = (int)ch; /* equivalent to: i = ch; i = i; */ i = (unsigned int)ch; /* equiv.: ui = ch; i = ui; */ i = (unsigned char)ch; /* equiv.: uch = ch; i = uch; */ There is one additional touch. A cast typically means: `I, the programmer, know that there is something slightly fishy about this assignment. I put the cast in to force a specific conversion sequence, so please do not warn about it.' This makes casts `more legal' than assignments in many cases, and makes them the only permitted (if implementation-defined) way of converting between integers and pointers, or different kinds of pointers. In article eichin@tweedledumber.cyGNUs.com (Mark Eichin) writes: >>The expression is (after macro expansion) >> switch ( (unsigned int) *(rpkt->dat+1) & ~1) { >>rpkt->dat is char*, it fetches the char into AL, then masks EAX and >>switches on that... leaving the upper 24 bits of EAX as garbage... so >>the switch always fails. This description conflicts slightly with itself. `Masking EAX' makes the upper 24 bits *not* garbage, if the upper 24 bits are masked away, but not if only the low order bit is masked away. I am not particularly conversant with Intel 80x86 assembly, but no sample compiler output appeared anyway. I suspect that the compilers in question make `char's signed, so this means: - Fetch 8 bits from rkpt->dat[1]. - Sign-extend the result to a full `int' (16 or 32 bits, depending on the compiler and/or model). - Convert that to `unsigned int' (no change in representation, but values that used to be negative are now considered to be large positive numbers). - Take away the low order bit. [output from sample program by Keith Smith, truncated radically by me, to show the value that results from a `char' that has been set initially to 0, 1, 126, 127, 128, and 255:] >0 / 0 >1 / 0 >126 / 126 >127 / 126 >128 / 4294967168 >255 / 4294967294 These results are exactly as expected on a two's complement 32-bit machine with signed `char'. 128 becomes -128, which is then interpeted in unsigned arithmetic as 2^32-128 or 4294967168. 255 becomes -1, or 2^32-1, or 4294967295; masked with &~1, it becomes 4294967294. The intention was almost certainly to do: switch (*(unsigned char *)(rpkt->dat+1) & ~1) or, equivalently: switch ((unsigned char)rpkt->dat[1] & ~1) An egregiously stupid compiler may generate slightly better code for the first version. Personally, I think the best solution might well be to change the type of the `dat' field of *rpkt to `unsigned char', so that the expression can be written as: switch (rpkt->dat[1] & ~1) but there may be other considerations that would make this inadvisable. -- In-Real-Life: Chris Torek, Lawrence Berkeley Lab CSE/EE (+1 510 486 5427) Berkeley, CA Domain: torek@ee.lbl.gov ------------------------------ From: axp@uniplex.co.uk (Alex Poon) Subject: SLS 1.03 and XS3 (solved)!! Date: 13 Aug 93 14:02:03 GMT Hello again, Just like to say thank you to all the people who replied to my problem. To get round all the redrawing problems you have to include the line vendor "nolinear" in your vga256 section. The XS3 server is now working faultlessly and amazingly it is faster than my Sun Sparc Station. Thanks again for all the help.... Alex +-------------------------------+------------------------------------------+ | Alex Poon (axp@uniplex.co.uk) | 486DX 33 PC + SLS 1.03 + XFree86 1.3 | | Software Engineer | Amiga A500 25MHz 030 + A590 | | Software Products Division | Nintendo Super Famicom | | Uniplex Limited UK | Disclaimer: My views are my own and | | | not those of Uniplex. | +-------------------------------+------------------------------------------+ ------------------------------ From: dale@mkseast.uucp (Dale Gass) Subject: Re: [Q] Diffs from SysV specs? Date: Mon, 16 Aug 1993 03:33:54 GMT kew@timesink.spk.wa.us (Keith Walker) writes: >>units - conversion program > Sometimes I really miss this program. Here's an entry I made to the Obfuscated C Contest that got turned down. It's a complete "units" program. I'll not include the unittab (since I haven't written a useful one), but it'll work with the standard Unix /usr/lib/unittab. I'm sure Linuxers can throw together a non-copyrighted unittab. I hereby put this program in the public domain, freely copyable. Just keep my credits, blah, blah, blah. (No CopyLeft's; go ahead and sell if you want to :-) Enjoy, -dale ==== /* Obfuscated "units" program, by Dale Gass, dale@mkseast.uucp */ #include #define N 999 int i=1,j,k,l,m,s;double c[N],e,w;FILE*f;char*p,*q,h,n[N][50],b[N],*r= "4mega A4kilo D4deci H5centiI5milliJ5microM4nano P4pico S";x(){fgets(b,N,f);} t(r){for(j=31;j<50;j++)if((k=n[m][j]*r)>0){printf("%c%s",l,n[j-30]);k>1&& putchar(k+'0');l='-';}}d(){printf("\t%e",c[m]);l=' ';t(1);l='/';t(-1);printf ("\n");}main(){f=fopen("/usr/lib/unittab","r");while(fscanf(f,"%s",n+i)>0)* n[i]=='/'?x():(a(),i++);f=stdin;for(;;){memset(n+i,0,100);if((printf( "you have: "),a())||(i++,printf("you want: "),a()))break;i--;memcmp(n[i]+31, n[i+1]+31,19)?(puts("conformability"),m=i,d(),m++,d()):printf( "\t* %e\n\t/ %e\n",c[i]/c[i+1],c[i+1]/c[i]);}}u(){e=j>0?10:(j=-j,.1);while (j--)w*=e;}v(){if((k=fscanf(f,"%lf%*[ \t]",&w))>0){j=fgetc(f);ungetc(j,f); if(j=='+'||j=='-'){fscanf(f,"%d",&j);u();}}}a(){g:v();c[i]=k?w:1;if(k<0)return (1);while((h=fgetc(f))=='|'){v();c[i]/=w;}if(s=h!='/')ungetc(h,f);s=2*s-1;while (1==(fscanf(f,"%*[ \t]"),fscanf(f,"%[^-/ \t\n]",q=b))){h=fgetc(f);if(q[0]=='*') {n[i][30]++;break;}p=strchr(q,0)-1;w=m=1;*p>='0'&&*p<='9'&&(m=*p-'0',*p=0);y: for(p=r;*p;p+=7){if(!strncmp(p+1,q,k=*p-'0')){q+=k;j=('G'-p[6])*s;u();goto y; }}z:for(j=1;j0?(c[i]*=c[j]):(c[i]/=c[j]);}c[i]*=w; }else{printf("cannot recognize %s\n",b);h!='\n'&&x();goto g;}if(h=='\n')break; h=='/'&&(s=-s);}return(feof(f));} ====== -- Dale Gass, Mortice Kern Systems, Atlantic Canada Branch Business: dale@east.mks.com, Pleasure: dale@mkseast.uucp ------------------------------ From: ronald@lowt.phys.tue.nl (Ronald Aarts) Crossposted-To: comp.os.linux.help Subject: [Q] Can't do networking Date: 16 Aug 1993 12:12:38 GMT I still have problems getting my network going with the latest SLS 1.03 release (disk a1.3 dated Aug 12). In doinstall I select option 5 for NFS-install and I specify my IP-number, network number, etc. My NFS-host is on the same subnet. It looks as if the network configuration is OK (no errors or warnings), but the NFS-mount doesn't work: ! mount clntudp_create: RPC: Port mapper failure - RPC: Unable to send My ifconfig and route look like: ! lo IP ADDR 131.155.110.55 BCAST 131.155.255.255 NETMASK 255.255.255.0 ! MTU 2000 METRIC 0 POINT-TO-POINT ADDR 0.0.0.0 ! FLAGS: 0x0049 ( UP LOOPBACK RUNNING ) ! ! eth0 IP ADDR 131.155.110.55 BCAST 131.155.255.255 NETMASK 255.255.255.0 ! MTU 1500 METRIC 0 POINT-TO-POINT ADDR 0.0.0.0 ! FLAGS: 0x0043 ( UP BROADCAST RUNNING ) ! ! Kernel routing table ! Destination net/address Gateway address Flags RefCnt Use Iface ! network * UH 0 43 eth0 ! softland * UH 0 0 lo I don't think the latter is correct !!! After that I installed Linux from my hard disk, but I still couldn't get the network going. ifconfig and route give similar output as above and NFS-mounts give the same error. I can ping to localhost, but I can't ping to any other host. From the other side, if I try to ping to my PC from e.g. an RS/6000 I don't get a reply, but now a host with IP-address 0.0.192.31 shows up in the arp-table of the RS ???? I also tried the hints of Darcy Boese in comp.os.linux/53085: /etc/ifconfig eth0 ${IPADDR} netmask ${NETMASK} broadcast ${BROADCAST} /etc/route add default gw ${GATEWAY} metric 1 /etc/route add ${IPADDR} /etc/ifconfig lo 127.0.0.1 /etc/route add 127.0.0.1 but now the first two route commands complain about "network unreachable". I'm using a 486 with 8 Meg RAM and a WD8013E ethernet card. I give up. Any hints ??????? -- Ronald Aarts Eindhoven University of Technology | Internet: ronald@lowt.phys.tue.nl Physics Department, P.O. Box 513, | Phone: +.31.40.474213 5600 MB Eindhoven, The Netherlands | Fax: +.31.40.453587 ------------------------------ From: floyd@hayes.ims.alaska.edu (Floyd Davidson) Subject: Re: Modem and noise problems - Please help Date: Mon, 16 Aug 1993 12:19:06 GMT Perhaps I should take this thread a little more serious. It should be in alt.dcom.telecom or comp.dcom.telecom, but on the chance that those who have posted do not have access to the appropriate groups I'm not going to redirect follow ups. I would suggest that anyone who has more questions or wishes to debate the subject should post to one of those groups and get this out of the linux group. (Email is fine too...) We have lost the attributes here, which may be just as well. I'm marking which statements I made: In article <24n8lh$em5@ursula.ee.pdx.edu> rkwee@ursula.ee.pdx.edu (Roland Kwee) writes: >floyd@hayes.ims.alaska.edu (Floyd Davidson) writes: >>>> >Otherwise, hook up an oscilloscope on the phone line and look for >>>> >disturbances. Or call your local phone company. This was my comment: >>>> O-scopes are nice things... but they are almost ALL made with an >>>> unbalanced input. If you connect that to the phone line, which >>>> is really a balanced transmission line, you will *cause* it to >>>> have all of the symtoms the original article described! >>>Most 'scopes have two inputs. If you use A+B (inverted) yu will have a >>>balanced, isolated input. ^. This was my comment: >>And if you know how to terminate it properly and can figure out >>what you are looking at... you probably have access to proper >>test equipment and wouldn't bother with a scope. >> >>O scopes are almost totally useless for testing telephone lines. >>I've only seen one person actually hook a scope to a cable pair, >>and when us old testboard techs sit around and tell jokes about >>the biggest bozo's we ever saw, thats the story I tell. Guys with >>a year or two of experience might get it, but guys with 25 or 30 >>years ALL find it quite funny. >> >>Want to tell me about how you can test cables with an O'scope? :-) And Roland Kwee comments: [some deleted] >About using an oscilloscope on the phone: the ****OBVIOUS**** solution >is to hook up the scope BEHIND the isolation transformer. That's the >point where the modem chip gets the signal, and what the modem chip >sees can the oscilloscope see too. Nothing magic. Doesn't take me >25 or 30 years to come up with. And if you think it is too hard to clip >the probes on the pins of that submicron PC board, open up a (junk) phone >set and measure there. No fancy equipment. That may be an "OBVIOUS" way to connect the scope, but then what do you do? Do you have any idea at all what you should see? Or what you shouldn't see? How are you going to know if what you see is good, bad, or indifferent? I have an extreme amount of experience in the use of scopes for circuit analysis and trouble shooting, and in analysis and testing of telephone lines. I can't interpret what you would see on a scope in any useful manner for testing telephone lines. And neither can you. The scope gives an instantaneous view of voltage vs. time, but that doesn't relate to the specifications for a telephone line. Neither noise nor signal levels can be measured with a scope. Put four equal level non-harmonically related tones on a telephone line and measure that with a properly balanced and terminated scope. Then remove two of the tones. What difference are you going to measure on the scope? The power level just changed 3 db, but the scope isn't going to tell you that. (It will... but if you are sharp enough to know how, I'll give you a problem where it will do exactly the same thing, and the level on the line *won't* change 3 dB on a proper test setup.) If you try measuring noise with a scope there is a little matter of spectral distribution (C message weighting) and peak to average ratios (a minimum of 12 dB, which is for gausian distribution). If your scope doesn't have facilities to handle that, it can't be used for accurate measurements. Telephone line parameters are not specified in terms that can be measured correctly with a scope. It might be very interesting to look at, but there is no way to relate what you see to the specifications for a telephone line. Floyd -- floyd@ims.alaska.edu A guest on the Institute of Marine Science computer Salcha, Alaska system at the University of Alaska at Fairbanks. ------------------------------ From: a59@aixterm1.urz.uni-heidelberg.de (Thurecht) Subject: multiusercard and driver searched !!! Date: Mon, 16 Aug 93 12:21:05 GMT Hi all, is there a working multiusercard (8) RS-422 and a driver under Linux?? Thanks Ernst ------------------------------ Crossposted-To: comp.unix.misc,comp.unix.sys5.r4 From: bill@bilver.uucp (Bill Vermillion) Subject: Re: Is this becoming comp.linux.advocacy? Date: Mon, 16 Aug 1993 12:28:00 GMT In article evan@telly.on.ca (Evan Leibovitch) writes: >There also happens to be a support issue involved. >Maybe the time will come when software vendors will explicitly report >Linux as being supported by its generic Intel binary. Just as Esix >eventually gained support as a product in its own right, so may Linux. >But make no mistake, being able to run the binary, on its own, is *not* >enough if it's not supported by the developer. And getting support could still be a problem. By being available as sources there can exist many variants, if it were not compiled completely stock. And it does seem to go through more variants than any commericial release. When having a problem with any supported product, if you are behind on release revs the typical reply is something like "so many things were fixed since xxx.xx you need to upgrade to see if this problem still exists. If you still have the problem after upgrading, please call back". How can you be sure the end user is using the, as you put it ' generic Intel binary ' ? -- Bill Vermillion - bill@bilver.uucp OR bill@bilver.oau.org ------------------------------ ** 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-Activists-Request@NEWS-DIGESTS.MIT.EDU You can send mail to the entire list (and comp.os.linux) via: Internet: Linux-Activists@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 tupac-amaru.informatik.rwth-aachen.de pub/msdos/replace The current version of Linux is 0.99pl9 released on April 23, 1993 End of Linux-Activists Digest ******************************