nsw@cord.UUCP (Neil Weinstock) (03/01/89)
[ If you can read this, there's something wrong with your line eater ] Recently there's been bundles of talk about DNet, Unix Windows, etc. Several people have asked if either program could be made to work with System V. I too would like something like that. The answer has generally been "System V doesn't have sockets, so it would be difficult." It seems, though, that people are missing the fact that System V has a multiplexed terminal protocol, called XT, built in, and a program "layers" that supports it. Why has no one ever attempted to do anything with XT? I haven't because it is far removed from any area in which I might fancy myself to have expertise. Offhand I can think of two approaches that might be fruitful: 1) Do something like DNet, with a host program that uses XT, and an Amiga front end that implements XT. 2) Write a 630 emulator for the Amiga, and let layers be the host end of the program. I can't see any reason why this wouldn't be possible, except perhaps that it might be difficult to obtain some of the necessary info from AT&T. (For those who don't know, the 630 is a bitmapped windowing terminal that runs with layers. I'm quite fond of mine, and I'd love to see the same functionality on my Amiga.) Option #2 is a bit more limited, since would just be a windowing terminal program, while #1 could be almost anything. Can anyone identify any major flaws in this idea, or is it simply that no one has gotten around to doing any of it? /.- -- .. --. .- .-. ..- .-.. . ... .- -- .. --. .- .-. ..- .-.. . ...\ / Neil Weinstock | att!cord!nsw | "One man's garbage is another \ \ AT&T Bell Labs | nsw@cord.att.com | man's prune danish." - Harv Laser / \.- -- .. --. .- .-. ..- .-.. . ... .- -- .. --. .- .-. ..- .-.. . .../
ditto@cbmvax.UUCP (Michael "Ford" Ditto) (03/02/89)
In article <723@cord.UUCP> nsw@cord.UUCP (Neil Weinstock) writes: >It seems, though, that people are missing the fact that System V has a >multiplexed terminal protocol, called XT, built in, and a program "layers" >that supports it. Why has no one ever attempted to do anything with XT? XT is not really much use for something like DNET because it only gives 8 virtual connections. Since that would be a silly limitation to put into DNET, it would still need its whole link protocol support anyway, so it's better to just do it over the raw serial line. Actually, I don't know what DNET uses sockets for, but I doubt that it's for the tty port multiplexing. >2) Write a 630 emulator for the Amiga, and let layers be the host end of the >program. I can't see any reason why this wouldn't be possible, except perhaps >that it might be difficult to obtain some of the necessary info from AT&T. This would be Truely Awesome, and has been on my "todo" list of Amiga hacks for a long time. Perhaps I can get some official blessing to dedicate some time to it now that Commodore is about to start selling a system with standard XT support (Amix has all the XT drivers and the "layers" program, etc.). Right now, an A2500UX with seven A500's hanging off it is pretty neat, but it would be much better if they could each have several Unix windows instead of just one. I'm pretty sure the info from AT&T can be had, but I think it might introduce some licensing problems (I wouldn't want said 630 emulator to be derived from AT&T code and therefore require a license fee to AT&T -- I'd rather build it just from available information.) Yes, it wouldn't be as general as DNET, but the host end is already written and standard with any recent SysV system. -- -=] Ford [=- "The number of Unix installations (In Real Life: Mike Ditto) has grown to 10, with more expected." ford@kenobi.cts.com - The Unix Programmer's Manual, ...!sdcsvax!crash!kenobi!ford 2nd Edition, June, 1972. ditto@cbmvax.commodore.com
hugh@censor.UUCP (Hugh D. Gamble) (03/02/89)
What I'd like to see (but not quite enough yet to get off my butt &
do it myself. Enough to pay for one though) is TCP/IP SLIP
support for the Amiga, then you can do lots of neat stuff & talk
to all the Joneses who run TCP/IP. (Can you say defacto standard :-)
--
Hugh D. Gamble (416) 581-4354 {lsuc, utzoo}!censor!hugh (Std. Disclaimers)
I don't want to live in a beer commercial,
I just want to play with some of the girls from one. :^)rminnich@super.ORG (Ronald G Minnich) (03/03/89)
In article <374@censor.UUCP> hugh@censor.UUCP (Hugh D. Gamble) writes: >What I'd like to see (but not quite enough yet to get off my butt & >do it myself. Enough to pay for one though) is TCP/IP SLIP >support for the Amiga, then you can do lots of neat stuff & talk >to all the Joneses who run TCP/IP. (Can you say defacto standard :-) Well, what can i say. I submitted this (for the fourth time in a year) to comp.sources.amiga some time ago. Just about that time the list froze up. I have a theory about comp.sources.amiga stewardship, and that is that it is a tiring enough job that it freezes the moderator up completely. The last time i tried to ship a version of Karn's tcp/ip with amiga mods was a year ago, and it still has not made it out! D*mn! Anyway the tcp/ip that may some year appear on comp.sources.amiga is nowhere near as good as Ameristar's, but it is fun to use. And a friend has put KA9Q on some very weird sys. v-like boxes, so there is no question that amigatcp can talk to sys v boxes. And, of course, amigatcp talks slip, which ameristar has not (yet ...) released. If only it could get distributed. sigh. ron P.S. for the adventurous the BSD tcp/ip is now freely redistributable (BUT NOT PUBLIC DOMAIN!)