ir320@sdcc6.UUCP (ir320) (11/25/85)
What are the advantages of being on different networks? Or to rephrase
that question:
o What hardware is required for UUCP,ARPA,BITNET,DECNET,MAILNET,CSNET?
(Would it be feasable to try to obtain a feed on one of these
networks. I have heard ARPA requires a very expensive controller
and special phone lines so it probably is not for most persons.)
o If anythig other than UUCP can be used, what are the advantages to
these other netowrks (forgot to include micnet what ever that is.)
John Antypas
UC San Diego
uucp: ...!{ucbvax,decvax,noscvax,ihnp4}!sdcsvax!lewak
arpa: lewak@sdcsvax.ARPA sdcsvax!lewak@Nosc.ARPA
other: lewak%sdcsvax.ARPA@{WISCVM.BITNET,DECWRL.DECNET,CSNET-RELAY.CSNET
MIT-MULTICS.MAILNET,XEROX}ron@brl-sem.ARPA (Ron Natalie <ron>) (11/27/85)
> o What hardware is required for UUCP,ARPA,BITNET,DECNET,MAILNET,CSNET? > (Would it be feasable to try to obtain a feed on one of these > networks. I have heard ARPA requires a very expensive controller > and special phone lines so it probably is not for most persons.) > > o If anythig other than UUCP can be used, what are the advantages to > these other netowrks (forgot to include micnet what ever that is.) > UUCP requires as a minimum a incoming modem. If you don't want to have to rely on someone else calling you to initiate a transfer, some sort of automatic dialer is required. The code is all in user mode and is available for nearly every UNIX system. Generally, except when it comes to NETNEWS, UUCP is very, very, cheap. ARPANET: Really the ARPA INTERNET. This is a collection of networks using a common protocol which automatically (at least to the user) routes the data to the required host on the required network through special hosts called Intenet Gateways. The per-host cost can be anything from a simple serial interface to an Ethernet board to something exotic like a Hyperchanel. It is pretty nice for campus collection of local area networks. The hard part is getting interfaced to the rest of the INTERNET. This requires some data line to some other part of the network. To connect directly to the ARPANET there are several different types of interfaces in use. DCA the people responsible for the hook ups these days are heading towards using an X.25 style interface, which would require a synchronous serial port and some pretty standard modems. This won't be ready until next summer however. Currently a specific serial protocl based on HDH is used over the same hardware. Companies like ACC make smart serial cards that do the X.25 and HDH parts for you. The hard part is getting authorization to get hooked up to the network. You need to either be contracted for by DARPA for research or have some DOD site speak up for you and say they require the ability to talk to you. Since in the past, hosts were frequently hooked up physically near the IMPs (these are the nodes on the ARPANET) there are two interfaces (really the same except for the line drivers) called local host and distant host. These are typically hooked up to the UNIBUS through ACC's LH/DH-11. The price for this is not any more than for a decent Ethernet card. The cute part was that ACC built a kludge box that allowed a host and IMP appear to be plugged into each other even though they were separated by a synchronous modem. This strategy was rather expensive, but for some of us, was the only way to do what we wanted. The software for this (TCP/IP) comes with the Berkeley UNIX's. It can be had (for a price) for System V and VMS. The ability to be on the INTERNET means instant interconnectivity with everyone else, had high data rates at least on the local group of networks. BITNET: BITNET is to the IBM world what UUCP is to the UNIX world. It also was a creature of neccessity. Based upon an interprocessor spooling system called RSCS, the hardware involved is just a serial line and a couple of modems. BITNET connections are pretty much dedicated links, and things are store and forwarded from host to host. Not quite as nice as being on the INTERNET but, pretty fast service. UNIX machines may be hooked up to BITNET by using the UREP package. I would say the main factor in deciding if you wish to be on BITNET is to decide if the people you really want to talk to are on BITNET. BITNET has some pretty far flung nodes throughout ACADEMIA in EUROPE and the MIDEAST. IBM uses a similar network between it's internal sites called VNET, but due to corporate paranoia this net is closed to all but specifically approved outside accesses. DECNET: DECNET is a network protocol/system not a specific network. You can have DECNET in your facility. There are corporate DECNETS that span cities, etc. DECNET can be used over a variety of network hardware from serial lines to Ethernet, etc. Pretty much DECNET provides the same functions as the ARPANET, but it is very, very specific to DEC hardware. In addition, the deciding factor is how to interconnect your DECNET to the rest of the world. DECNET support has been rumored for the next ULTRIX release. MAILNET: I'm not sure of this one, but it appears to be a collection of dial up hosts somehow gatewayed to the US through Multics. I just mail to them through there, rather than bothering worrying about being on MAILNET. CSNET: CSNET has two modes of operation. The first is over TCP/IP style connections involving the ARPANET and a TCP/IP over X.25 project. The other older mode of operation is a dial up protocol using Dave Crocker's Phonenet protocol (please, don't say MMDF, MMDF is the mail queing system and is separate form the Phonenet protocol, you can use either without the other). The CSNET relay machine calls you up and transfers mail to you. In addition, CSNET has several server functions such as a white pages directory. Of course, this all isn't free, it costs to be on CSNET, but it does give you a good working and legal connection to the ARPANET for mail at least. -Ron
avolio@decuac.UUCP (Frederick M. Avolio) (11/27/85)
In article <560@brl-sem.ARPA>, ron@brl-sem.ARPA (Ron Natalie <ron>) writes: > DECNET: DECNET is a network protocol/system not a specific network. > ... DECNET support has been rumored for the next ULTRIX release. DECnet support is in the *present* release of Ultrix-32 (V1.1). -- Fred @ DEC Ultrix Applications Center {decvax,seismo,cbosgd}!decuac!avolio
spw2562@ritcv.UUCP (11/27/85)
In article <2301@sdcc6.UUCP> ir320@sdcc6.UUCP (ir320) writes: > o What hardware is required for UUCP,ARPA,BITNET,DECNET,MAILNET,CSNET? > > o If anythig other than UUCP can be used, what are the advantages to > these other netowrks (forgot to include micnet what ever that is.) > >John Antypas >UC San Diego BITNET uses leased lines, but I don't know what hardware is neccesary. Because it has leased lines, communication between any two nodes is (nearly) instantaneous, or more accurately, you seldom have to wait more than several seconds for a message to get to it's destination. In real extreme cases, you may have to wait several minutes, or if a node is down, you could wait indefinetely. Routing is done automatically, you just specify a destination node, and user, if necessary. However, there is no organized news network like Usenet's. There is (usually) only one allowable path between any two nodes. The method of routing is a table at each node which basically consists of two entries: the destination node, and the adjacent node which messeges for that destination are passed through. The problem with this is that when a new node is added, it may take quite a while for updates to permeate(sp?) the net. I've had at least one instance where I could send messages to one node on bitnet, but they could not send back to my, because a node between us had a routing for them, but not for the node I was on. It was a royal pain because there were some files I desperately wanted at that node. Hope this helps.. ============================================================================== Steve Wall [Snoopy] @ Rochester Institute of Technology USnail: 6675 Crosby Rd, Lockport, NY 14094, USA Usenet: ...!ritcv!spw2562 Unix 4.2 BSD BITNET: SPW2562@RITVAXC VAX/VMS 4.2 Voice: Yell "Hey Steve!" Disclaimer: What I just said may or may not have anything to do with what I was actually thinking...
spw2562@ritcv.UUCP (11/27/85)
[ There's no such thing as a linnmmpphh ] *munch munch* Incidentaly, The network commonly called BITNET is really three networks.. BITNET (US), NORTHNET (Canada), and CERN (Europe). (I think).
spw2562@ritcv.UUCP (Snoopy) (11/29/85)
In article <560@brl-sem.ARPA> ron@brl-sem.ARPA (Ron Natalie <ron>) writes: >BITNET: BITNET is to the IBM world what UUCP is to the UNIX world. Last I heard, 50%+ of the sites on BITNET where VMS ( and Unix? ). VMS is typically interfaced via the JNET package, don't know if there are other packages available. >It also was a creature of neccessity. Based upon an interprocessor >spooling system called RSCS, the hardware involved is just a serial >line and a couple of modems. BITNET connections are pretty much dedicated >links, and things are store and forwarded from host to host. As I said earlier, BITNET uses leased lines, meaning totally dedicated. >UNIX machines may be hooked up to BITNET by using the UREP package. Also, BERK, and several other packages. ======================================================================== Steve Wall [Snoopy] @ Rochester Institute of Technology USnail: 6675 Crosby Rd, Lockport, NY 14094, USA Usenet: ...!ritcv!spw2562 Unix 4.2 BSD BITNET: SPW2562@RITVAXC VAX/VMS 4.2 Voice: Yell "Hey Steve!" Disclaimer: What I just said may or may not have anything to do with what I was actually thinking...
hes@ecsvax.UUCP (Henry Schaffer) (11/30/85)
> > BITNET uses leased lines, but I don't know what hardware is neccesary. > Because it has leased lines, communication between any two nodes is > (nearly) instantaneous, or more accurately, you seldom have to wait more > than several seconds for a message to get to it's destination. In real > extreme cases, you may have to wait several minutes, or if a node is down, > you could wait indefinetely. > USnail: 6675 Crosby Rd, Lockport, NY 14094, USA The hardware needed is a modem on each end of the leased line, and then the normal computer hook up to the modem. A comment on performance - it certainly can be this good when there is little traffic. However the delay can be *much* more than seconds-minutes for a file transfer when there are many other files ahead of you. (I think that short e-mail messages are given priority over large files, so that message transfer does work very well in practice.) --henry schaffer
phil@amdcad.UUCP (Phil Ngai) (12/02/85)
In article <817@ecsvax.UUCP> hes@ecsvax.UUCP (Henry Schaffer) writes: >> BITNET uses leased lines, but I don't know what hardware is neccesary. >> Because it has leased lines, communication between any two nodes is >> (nearly) instantaneous, or more accurately, you seldom have to wait more > > A comment on performance - it certainly can be this good when there is >little traffic. However the delay can be *much* more than seconds-minutes >for a file transfer when there are many other files ahead of you. (I think >that short e-mail messages are given priority over large files, so that >message transfer does work very well in practice.) If BITNET uses the RSCS stuff, I believe it has the characteristic of transfering only one file over one physical link at a time so that if your e-mail message gets stuck behind a huge 20 Mb file transfer in progress the short msg has to wait for the long file to finish. Transferring short msgs before long msgs are one reason USENET articles can arrive out of sequence. (totally unrelated to the previous paragraph, just something I thought I'd toss in) -- There's nothing I hate more than sorting socks. Phil Ngai +1 408 749-5720 UUCP: {ucbvax,decwrl,ihnp4,allegra}!amdcad!phil ARPA: amdcad!phil@decwrl.dec.com
dmcanzi@watdcsu.UUCP (David Canzi) (12/02/85)
In article <9080@ritcv.UUCP> spw2562@ritcv.UUCP (Snoopy) writes: >Incidentaly, The network commonly called BITNET is really three networks.. >BITNET (US), NORTHNET (Canada), and CERN (Europe). (I think). Not quite, they're called BITNET, NETNORTH, and EARN. (Some people wanted to call the Canadian network "GWNET" (Great White NET), but the humourless bureaucratic types prevented it.) They *really* *are* one network, but there is a common illusion that the political boundaries between parts of the network are important. -- David Canzi "But lo! men have become the tools of their tools." -- Henry David Thoreau
spw2562@ritcv.UUCP (Fishhook) (12/03/85)
In article <1933@watdcsu.UUCP> dmcanzi@watdcsu.UUCP (David Canzi) writes: >In article <9080@ritcv.UUCP> spw2562@ritcv.UUCP (Fishhook) writes: >>BITNET (US), NORTHNET (Canada), and CERN (Europe). (I think). >Not quite, they're called BITNET, NETNORTH, and EARN. (Some people >-- >David Canzi I stand corrected. Thanks Dave. ============================================================================== Steve Wall @ Rochester Institute of Technology USnail: 6675 Crosby Rd, Lockport, NY 14094, USA Usenet: ..!rochester!ritcv!spw2562 (Fishhook) Unix 4.2 BSD BITNET: SPW2562@RITVAXC (Snoopy) VAX/VMS 4.2 Voice: Yell "Hey Steve!" Disclaimer: What I just said may or may not have anything to do with what I was actually thinking...
herbie@polaris.UUCP (Herb Chong) (12/04/85)
In article <7018@amdcad.UUCP> phil@amdcad.UUCP (Phil Ngai) writes: >If BITNET uses the RSCS stuff, I believe it has the characteristic of >transfering only one file over one physical link at a time so that if >your e-mail message gets stuck behind a huge 20 Mb file transfer in >progress the short msg has to wait for the long file to finish. once RSCS has begun transmitting a file, it does not stop. when selecting a file for transmission though, it chooses the shortest highest priority file first. messages can be transmitted at the same time as a file transfer but are limited in length. Herb Chong... I'm still user-friendly -- I don't byte, I nybble.... VNET,BITNET,NETNORTH,EARN: HERBIE AT YKTVMH UUCP: {allegra|cbosgd|cmcl2|decvax|ihnp4|seismo}!philabs!polaris!herbie CSNET: herbie.yktvmh@ibm-sj.csnet ARPA: herbie.yktvmh.ibm-sj.csnet@csnet-relay.arpa ======================================================================== DISCLAIMER: what you just read was produced by pouring lukewarm tea for 42 seconds onto 9 people chained to 6 Ouiji boards.