hays@apollo.UUCP (03/03/87)
CROSS POSTED TO MANY AREAS FOR CROSS FERTILIZATION:
I have been following the MINIX group now from it's inception and
would like to suggest the following two items be considered for
STANDARD EXTENSIONS:
1. MIT's X Window System - for those with bitmaped screens
this would provide a standard windowing system for applications with
a migration path to/from larger systems. (Supported Standard: Apollo,
HP, DEC, DG, Adobe, others [assume trademark notices]).
2. KA9Q's TCP/IP (Phil Karn) implementation. This code,
which Phil will share for NON-COMMERCIAL reproduction, implements
a driver for an Ethernet Controller as well as serial line, HDLC,
and Amateur Packet Radio connections.
(*FOR U.S.A, other countries may have different requirements*)
For those who are interested: Starting in March the lowest
class of Amateur License will permit PACKET RADIO priviledges and
HOBBY computerists should consider getting licenced as with Phil's
code and the available frequencies it would be possible to build
a network for sharing BINARIES and SOURCES for HOBBY MINIX programs,
as well as non-commercial electronic mail.
This lowest class license [NOVICE] will require you to
pass the following tests:
1. International Morse Code (receive only) at 5 WPM
(3-40 hours of study). You listen to 5 minutes
(25 Words/125 characters) of plain text and answer
a 10 question test on content. (Answers must be
letter perfect)
2. Theory and Law - 30 multiple choice questions on
OHMs Law, Legal Frequencies, etc. VERY SIMPLE
(Read a small tutorial).
If you have questions about licensing, etc. on ham
radio, POST to rec.ham-radio on USENET.
John Hays
Amateur Radio Station: KD7UW
--
John D. Hays, Consultant UUCP: ...!decvax!wanginst!apollo!hays
Corporate Systems Engineering ...!uw-beaver!apollo!hays
Apollo Computer Inc. CIS: 72725,424 {weekly}
!MY OPINIONS, NOT Apollo's!mjranum@gouldsd.UUCP (03/03/87)
In article <336daeb9.9540@apollo.uucp>, hays@apollo.uucp (John Hays) writes: > would like to suggest the following two items be considered for > STANDARD EXTENSIONS: > > 1. MIT's X Window System - for those with bitmaped screens > > 2. KA9Q's TCP/IP (Phil Karn) implementation. This code, [blah,blah,blah...] Let's be real, here, guys. From what I understand (not having had a chance to get my minix up, running, and beaten on) there are even questions as to whether the hard disk drivers function properly. You are talking about implementing an awful lot of fancy stuff on some pretty brain-damaged hardware. I am sick of seeing all this 'when I get suntools running on minix I'll post to the net" shit. Don't waste the net's time telling us about the great things you're going to do unless you have a beta-test version :-) I think it might be more productive to worry (in order) about exhaustively testing minix, and making more portable drivers for a variety of hard disk controllers, and then maybe clearing up the odd bug here and there. TCP/IP is really very nice, but who wants to talk to a PC anyway ? I'd never let one on my network unless the sucker had a better security system than being able to boot off of *ANY* floppy you choose (hence any /etc/passwd, etc). Please wake up and smell the roses, guys. UNIX wasn't written overnight, and it certainly wasn't written by 800 usenet messages saying "well, I'd really like to see this implemented..." or "I plan to port all the code for KERMIT and build it into the kernel as a local area network", blah, blah, blah... I suggest we maybe form a talk.minix.wild.wetdream, or a comp.os.fantasy for you guys, and the rest of us can concentrate on simpler things like maybe making it a bit more portable, robust, and adding some of our favorite AT&T-like tools. Ferget the TCP/IP - the basics like 'sed' and 'lex' are a lot more likely to be missed. --mjr(); -- "It is better to shred the bugger than to bugger the shredder." -ancient doltic proverb.
madd@bucsb.bu.edu.UUCP (03/04/87)
In article <475@gouldsd.UUCP> mjranum@gouldsd.UUCP (Marcus J Ranum) writes: >I think it might be more productive to worry (in order) about exhaustively >testing minix, and making more portable drivers for a variety of hard disk >controllers, and then maybe clearing up the odd bug here and there. TCP/IP >is really very nice, but who wants to talk to a PC anyway ? I'd never let >one on my network unless the sucker had a better security system than being >able to boot off of *ANY* floppy you choose (hence any /etc/passwd, etc). THERE IS NOT ONE SINGLE UNIX SYSTEM ANYWHERE THAT IS INVULNERABLE TO BOOTING DOCTORED MATERIAL. If you can set up the system in the first place, then you can also set it up again with altered files. The only thing that stops most people from doing this is a locked door. You can just as effectively lock up a PC. Therefore, I don't think that the ability to boot off a diskette is a valid reason not to allow PC's to connect to a network. Now that that's off my chest, the rest of the arguments given in M. J. Ranum's posting are great. Let's not try to set up complex programs on MINIX systems until MINIX is humming along perfectly. %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% - Jim Frost * The Madd Hacker - UUCP: ..!harvard!bu-cs!bucsb!madd | ARPANET: madd@bucsb.bu.edu CSNET: madd%bucsb@bu-cs | BITNET: cscc71c@bostonu -------------------------------+---+------------------------------------ "Oh beer, oh beer." -- Me | [=(BEER) <- Bud the Beer (cheers!)
bdale@winfree.UUCP (Bdale Garbee) (03/08/87)
In article <475@gouldsd.UUCP> mjranum@gouldsd.UUCP (Marcus J Ranum) writes: >...concentrate on simpler things like >maybe making it a bit more portable, robust, and adding some of our favorite >AT&T-like tools. Ferget the TCP/IP - the basics like 'sed' and 'lex' are >a lot more likely to be missed. I suppose it's all a matter of your perspective. Phil and I (and the other folks who've been helping beat on our code) have a working TCP/IP package running under MS-Dos. Neither of us like MS-Dos. We see Minix as a neat next step for the package, in that it will allow us to "do things right" without having to write an OS in the process. A full blown unix system with TCP/IP isn't our goal. I do indeed hope that someone works on sed, and lex... a driver that will talk to other hard disk controller cards, and maybe even (oh gosh... here's a REAL wet dream :-) an RS-232 driver and uucp. In the meantime, some of us will be working on TCP/IP. -- Bdale Garbee, N3EUA phone: 303/593-9828 h, 303/590-2868 w uucp: {bellcore,crash,hp-lsd,hpcsma,ncc,pitt,usafa,vixie}!winfree!bdale fido: sysop of 128/19 packet: n3eua @ k0hoa, Colorado Springs
jdmorrison@watdragon.UUCP (03/10/87)
In article <230@winfree.UUCP> bdale@winfree.UUCP (Bdale Garbee) writes: >to other hard disk controller cards, and maybe even (oh gosh... here's a REAL >wet dream :-) an RS-232 driver and uucp. Does this mean that stock Minix does not know how to use the serial port if there is one on the PC? Just wondering how much trouble it would be to write a terminal emulator, etc... -- Joe Morrison VLSI Group University of Waterloo {decvax,allegra,ihnp4}!watmath!watvlsi!joe -- "The first fifteen million years -- they were the worst..."