jr@wlcrjs.UUCP (Jim Rosenberg) (11/22/84)
By the process of consent by osmosis, the Venix Users Group is hereby christened VUG. I've heard from several people via usenet & several more have left their names on the database at Unisource. In cooperation with them I'm going ahead to try to organize the group. Pending some calamity I now plan to be at UniForum where I will try to organize a B.O.F. session where we can get together. NOTE: If you left your name on the uni3 machine at Unisource as being interested in a VUG and are on Usenet, please send me your net address. (uucp mail preferably.) I've heard from several users of Venix on PDP-11ish machines & we should probably form a single group, or else coordinate our efforts. /VUG/motd: 1. Dorris at Unisource tells me that Unify is now available under Venix. Leverage sounds like an interesting product but as I understand it it's not a true relational DBMS, so those interested in running a DBMS under Venix should be grateful for this. Unisource *is* interested in marketing competing applications in the same niche. They'd love to market everything they can get their hands on, but have had problems getting the vendors to port to Venix. Any ideas on making this easier would be welcome. Better memory models for the C compiler would obviously help. I believe 2.0 has a medium memory model; some programs might need a large model to be ported. 2. For those wanting an Xmodem program for Venix I have found the best of the bunch to be uc, downloaded from /vaxsig/unix XA on CompuServe. Believe it or not it compiles on both 4.1 BSD and Venix with *no ifdefs*! Let me know if you need the code. 3. Venturcom has *put out an appeal* to Unisource to come up with a wish list for System V. This is an offer we won't refuse!! It might be difficult to prioritize our wishes through usenet -- maybe we can do this at UniForum -- but for starters I will be delighted to pass on all requests for what we want in the Sys. V. release. My own personal wish list follows, in no particular order: A. Clean up tty read() in raw mode to make it consistent with other flavors of Unix & return when 1 char arrives from a multi-char. read. B. More full suite of Berkeley stuff, including: Mail, ctags, strings, xstr, mkstr, etc. C. 4.2 BSD version of uucp utitilizing modemcap, if possible. (This is not a pipe dream. This version of uucp is available under Sys. V on the Convergent Mini-Frame.) If that's not possible do *something* to uucico so that the timeout does not abort the dial on Hayes-compatible modems. D. Korn shell if AT&T will release it. E. Better flexibility on FDISK partitions. I've talked to at least two people who paid for Venix & are not using it because they need a partition for DOS and one for Concurrent. As it comes off the shelf I think Venix can be configured to run in only two partitions, by simply not mounting /tmp. But there's no documentation on this for those not familiar with Unix. 4. We're going to try to coordinate a single unified bug list. If you have a bug list & haven't sent it to me I'd appreciate a copy. Thanks to Randy Seuss at ihnp4!wlcrjs!randy for putting up his machine as a bbs for net news. It's easiest to get in touch with me by uucp mail at: Jim Rosenberg decvax!idis!pitt!amanuen!jr (412) 784 2806
guy@rlgvax.UUCP (Guy Harris) (11/27/84)
> 3. Venturcom has *put out an appeal* to Unisource to come up with a wish list > for System V. ... My own personal wish list follows, in no particular > order: > > A. Clean up tty read() in raw mode to make it consistent with other flavors > of Unix & return when 1 char arrives from a multi-char. read. Not applicable to System V. There's no such thing as "raw" mode in S5; there's a bunch of modes you turn on and off. In "~icanon" mode (which turns off erase and kill processing), you can say how many characters should complete a multi-character read, and can also say that the read should complete in some number of 10ths of a second *if* at least one character has come in. (NOTE: some people have thought that this is a read timeout, and that the clock starts running at the time the "read" is issued. It just ain't so. In S5, if the minimum number of characters to complete the read is 0, the clock does start when the "read" is issued, but only then; this is emphatically *not* the case with S3, however.) > B. More full suite of Berkeley stuff, including: Mail, ctags, strings, xstr, > mkstr, etc. Well, with S5 Release 2 you get "Mail", only Bell renamed it "mailx" and changed it around a bit. No credit given, from what I can tell, except in the comments in the source code. Humph. Guy Harris {seismo,ihnp4,allegra}!rlgvax!guy
Goeke@MIT-MULTICS.ARPA (11/28/84)
Do add me to the list of those interested in a VENIX users
group. I've been running it on an 11/23 with a 20MB DSD drive for a
year and a half now, and have been very happy with it. Interestingly,
Venturcom seems to have forgotten that I exist, since absolutely ZERO
mail has arrived on new products, etc. since the day I got my floppies.
They share this distinction with DEC who also doesn't realize that past
customers with licenses are possible future customers.
BTW, I communicate to the rest of the world through this
Honeywell machine called MIT-Multics for a couple of reasons. Firstly,
I've never gotten 'uucp' to work, though I only spent a couple of hours
trying. Secondly, a true net link would mean I'd have to connect my
machine to a telephone. Now in real life, you should understand, my
machine works without passwords. . .if you can get to one of it's two
terminals, you get on -- the same with my desk and file cabinet. I'm
less happy with having to institute some security scheme just because I
want a net connection occasionaly. Anyway, I use Kermit to talk to
MIT-Multics and drop files; works like a champ, and it compiled with no
difficulties; I may have had to set 1ifdef, but that was it.
-- Bob Goeke
MIT Center for Space Research
Room 37-567
Cambridge, MA 02139
617-253-1910rosen@siemens.UUCP (11/28/84)
Just a bit of curiosity... What is Venix and what makes it so
special and popular that you want to start your own group?
Steve Rosen
Siemens RTL
{princeton | eosp1}!siemens!gypsy!rosenhenry@utzoo.UUCP (Henry Spencer) (11/30/84)
> Well, with S5 Release 2 you get "Mail", only Bell renamed it "mailx" and > changed it around a bit. No credit given, from what I can tell, except in > the comments in the source code. Humph. Well, that's better than Berkeley usually does. You won't even find comments in the source to indicate that the original vtroff software came from U of Toronto. Which is an out-and-out violation of U of T's software licence, which requires credit in all re-use. -- Henry Spencer @ U of Toronto Zoology {allegra,ihnp4,linus,decvax}!utzoo!henry
mark@cbosgd.UUCP (Mark Horton) (12/01/84)
In article <4700@utzoo.UUCP> henry@utzoo.UUCP (Henry Spencer) writes: >Well, that's better than Berkeley usually does. You won't even find >comments in the source to indicate that the original vtroff software >came from U of Toronto. Which is an out-and-out violation of U of T's >software licence, which requires credit in all re-use. >-- > Henry Spencer @ U of Toronto Zoology > {allegra,ihnp4,linus,decvax}!utzoo!henry This is probably my fault. I wrote the Berkeley Font Catalog and did a lot of the font work. I probably should have credited Toronto as the source of the original software and the Hershey fonts. (I think in turn the fonts came from the USA dept of something or other.) What happened is that I never saw a license from Toronto for that stuff, and until I read this netnews article I didn't know such a license existed. I was not the original recipient of the stuff from Toronto - I'm not sure who was. In any case, it just never occurred to me to insert an acknowledgement to Toronto, or to MIT and Stanford from whence most of the rest of the fonts came. Things were different back then, you just passed software around and didn't worry too much about where it came from. So I hereby publicly apologise to the University of Toronto for not crediting them as the original source of the vtroff software. Mark Horton