Another network protocol I wanted to exhibit someday was IBM's Token Ring. - Bill Degnan
A few S-100 companies supported ARCnet. Boards were available from Compupro and a few others companies. I've collected a few ARCnet devices and S-100 cards. But this stuff is outside my digital EE experience. And if I got one of these cards running, what would it talk to? Other than my showing interest, my point is that there may be interest from INexperienced persons who own vintage computers which MIGHT operate on these networks, for purposes of an exhibit or a repair workshop. Those persons (me) would need some handholding and known-good network traffic. So a consideration for a VCF-East exhibit, would be to string cabling behind several exhibit-tables; rather than one table with these puppies sitting next to each other. I think a "grand exhibit" is a lot of work and coordination; whereas just connecting adjacent exhibits requires less planning. Similar thoughts for any VCF or other-site workshop. Another consideration, as I mentioned is "whom do I talk to?" That MIGHT suggest, construction of some modern device (Arduino or RaspPi based) to act as a peer. NOthing fancy, just enough to validate a *real* networking device. While some of us have several node-computers to toy with, many individuals may only have one computer or one kind or one network interface. Of course Arduinos and RAspPis are common. So I think, if a handful of people say they are prepared to support a particularly obscure networking protocol, and are willing to show up a few times to help others become "nodes", and they can point to some resources like I suggested - some kind of workshops or exhibits could be done. A year out from VCF is *reasonable* to gather such interest, as it may take time to gather or buy resources and make them work, and then make them work *together*. I've just provided a strategy - who has the interest and hardware? For instance Bill Degnan has found several people are interested and have supported by trade, Novell Netware. Even I have a few Lantastic network cards for IBM-PC (saved because of the Z180 chip on them). Are they ARCnet? I dunno....? But I think Bill Degnan may have a deal, if those who responded wanna work together. Herb "no net" Johnson -- Herbert R. Johnson, New Jersey in the USA http://www.retrotechnology.com OR .net
On 05/25/2018 03:46 PM, Herb Johnson via vcf-midatlantic wrote:
A few S-100 companies supported ARCnet. Boards were available from Compupro and a few others companies. I've collected a few ARCnet devices and S-100 cards. But this stuff is outside my digital EE experience. And if I got one of these cards running, what would it talk to?
As of a few years ago anyway, there was still ARCnet support in Linux. A friend of mine (one of the Digex guys) once ran an ARCnet subnet in his house, using a Linux box to bridge Ethernet to ARCnet. But then you'd be at the point of implementing protocols. It can surely be done, given sufficient motivation. And it would be very, very cool. -Dave -- Dave McGuire, AK4HZ New Kensington, PA
Another consideration, as I mentioned is "whom do I talk to?" That MIGHT suggest, construction of some modern device (Arduino or RaspPi based) to act as a peer. NOthing fancy, just enough to validate a *real* networking device. While some of us have several node-computers to toy with, many individuals
I made a joke to someone at VCF who had built a similar device to the wifi232 that he should make it do AppleTalk to some sort of massive AppleTalk Cloud server that connects tons of people together for file sharing. Kind of like a BBS? Not sure how AppleTalk permissions work though. Printer emulation even! Could print to pdf.
Itg's unlikely that a modern Ethernet or WiFi device, could perform the hardware and software protocols corresponding to even older networking standards, including Appletalk. I think such emulators would have to be built from the ground up. They are likely painful to build reliably and flexibly. Remember my concept - sufficient to "echo", handshake some packets - not a full emulation. That said, I happen to know, AppleTalk did run "over Ethernet", through simple devices that physically routed AppleTalk traffic on the Ethernet wires. But the Ethernet devices themselves, ignored that "traffic" - probably saw it as bad packets, noise, etc. This was done to put Apple printers on Ethernet networks - except of course any Ethernet-owning device had zero access to Apple printers, and Apple computers had zero access to Ethernet devices. People who buy Appletalk printer to Ethernet devices, thinking they are their magic road to "internet their Macs", find out eventually that won't happen. So it may be possible, for a few network protocols to share the same cable. But they would not be "interoperative" - that calls for hard work. Companies and careers were made, doing such things in the era. We poor hobbyists live in the shadows of such giants. The amazing thing, would be to get some ancient networking protocol to work AT ALL. I suggest humility, and avoid the simple notion "I can plug it in - so it will work, or could". Herb On 5/25/2018 3:54 PM, Ethan O'Toole wrote:
Another consideration, as I mentioned is "whom do I talk to?" That MIGHT suggest, construction of some modern device (Arduino or RaspPi based) to act as a peer. NOthing fancy, just enough to validate a *real* networking device. While some of us have several node-computers to toy with, many individuals
I made a joke to someone at VCF who had built a similar device to the wifi232 that he should make it do AppleTalk to some sort of massive AppleTalk Cloud server that connects tons of people together for file sharing. Kind of like a BBS? Not sure how AppleTalk permissions work though.
Printer emulation even! Could print to pdf.
-- Herbert R. Johnson, New Jersey in the USA http://www.retrotechnology.com OR .net
On 05/25/2018 06:39 PM, Herb Johnson via vcf-midatlantic wrote:
Itg's unlikely that a modern Ethernet or WiFi device, could perform the hardware and software protocols corresponding to even older networking standards, including Appletalk. I think such emulators would have to be built from the ground up. They are likely painful to build reliably and flexibly. Remember my concept - sufficient to "echo", handshake some packets - not a full emulation.
That said, I happen to know, AppleTalk did run "over Ethernet", through simple devices that physically routed AppleTalk traffic on the Ethernet wires. But the Ethernet devices themselves, ignored that "traffic" - probably saw it as bad packets, noise, etc. This was done to put Apple printers on Ethernet networks - except of course any Ethernet-owning device had zero access to Apple printers, and Apple computers had zero access to Ethernet devices.
People who buy Appletalk printer to Ethernet devices, thinking they are their magic road to "internet their Macs", find out eventually that won't happen.
So it may be possible, for a few network protocols to share the same cable. But they would not be "interoperative" - that calls for hard work. Companies and careers were made, doing such things in the era. We poor hobbyists live in the shadows of such giants. The amazing thing, would be to get some ancient networking protocol to work AT ALL. I suggest humility, and avoid the simple notion "I can plug it in - so it will work, or could".
I want to throw out a caveat here. My 8 bits days were not networked other than a modem. When I got into networking in the late 80's there were lots to chose from. This was the era of the PC so most of my background comes from getting PCs to talk to other things like Mini's and mainframes. Ah now you're in my wheelhouse (well kind of). I still have a Cisco router that can speak most of these protocols. I think Dave was interested in these devices (the CGS and the ACS). As Herb noted Appletalk could be bridged to Ethertalk but didn't talk to TCP/IP. This was taking the data layer (?) protocol and changing layer 1 (ISO state). Other methods of networking were RS485 (full duplex and half duplex). I'm not sure many microcomputers used this or other networking protocols but I know a few business microcomputers did (Digicomp? Digi something?). I know Atari attempted to use their SIO (Serial I/O) bus to do a star. I know that my Circuit Cellar HCS II used RS485 half duplex (master/slave). I recall an IBM PC with a 3Com card in it (I think the card had a 68000 processor on it). The PC ran at 4.77 MHz and the 68K ran at 8MHz. I think the network protocol was TCP/IP. As herb said getting the protocols to talk to each other is a parin. Get the layer one to work together was no bad as long as you stuck to 802.3 (thin, thick, or 10BaseT). You could probably get 100baseT and above to work but performace could get interesting. For many of us from the 8 bit era, the modem and a BBS was the general method of networking followed by sneaker-net. Or: Never underestimate the bandwidth of a station wagon full of tapes hurtling down the highway. (Andrew S. Tanenbaum paraphrasing Dr. Warren Jackson) -- Linux Home Automation Neil Cherry ncherry@linuxha.com http://www.linuxha.com/ Main site http://linuxha.blogspot.com/ My HA Blog Author of: Linux Smart Homes For Dummies
On 05/25/2018 09:17 PM, W2HX wrote:
UUCP!
Actually, very true! -- Linux Home Automation Neil Cherry ncherry@linuxha.com http://www.linuxha.com/ Main site http://linuxha.blogspot.com/ My HA Blog Author of: Linux Smart Homes For Dummies
I've actually been working on some options for routing LocalTalk to Ethernet (including a MacIP gateway). The bottom line is that most modern WiFi/Ethernet devices can usually handle the protocols that go over Ethernet just fine (with some exceptions; there are features of some protocols that work much better over Ethernet than WiFi, particularly if they do funny things with MAC addresses). Serial protocols are another matter. Ones that run over standard UARTs, like DECNet, are totally doable with a lot of new things. LocalTalk, on the other hand, made use of the SDLC mode in the 8530 serial chip used on the Macintosh, which is not something generally present on a lot of embedded devices. At that point, you're left with the options of interfacing an external 8530 (messy), bit-banging the serial (tough at the 230 Kbps used for LocalTalk, but sometimes doable), or making your own translator chip from a CPLD or FPGA (expensive). One such project I've been working on is with the BeagleBone Black, which has two high-speed real-time coprocessors on the chip; those are quite suited to bit-banging the LocalTalk serial and passing it to the main CPU. All of that, of course, comes before the software stack. The benefit of something like the BBB is that it runs Linux, so it can run netatalk, which already routes quite well between LocalTalk and EtherTalk (and can do the MacIP gateway as well). For something like an ESP32, you'd need to write your own embedded AppleTalk stack, which is not trivial. FWIW, if you're looking to connect "AppleTalk to cloud", check out the forums at mac68k.info; Rob Braun has written what is essentially an AppleTalk VPN module for doing essentially that. - Dave
On May 25, 2018, at 18:39, Herb Johnson via vcf-midatlantic <vcf-midatlantic@lists.vintagecomputerfederation.org> wrote:
Itg's unlikely that a modern Ethernet or WiFi device, could perform the hardware and software protocols corresponding to even older networking standards, including Appletalk. I think such emulators would have to be built from the ground up. They are likely painful to build reliably and flexibly. Remember my concept - sufficient to "echo", handshake some packets - not a full emulation.
That said, I happen to know, AppleTalk did run "over Ethernet", through simple devices that physically routed AppleTalk traffic on the Ethernet wires. But the Ethernet devices themselves, ignored that "traffic" - probably saw it as bad packets, noise, etc. This was done to put Apple printers on Ethernet networks - except of course any Ethernet-owning device had zero access to Apple printers, and Apple computers had zero access to Ethernet devices.
People who buy Appletalk printer to Ethernet devices, thinking they are their magic road to "internet their Macs", find out eventually that won't happen.
So it may be possible, for a few network protocols to share the same cable. But they would not be "interoperative" - that calls for hard work. Companies and careers were made, doing such things in the era. We poor hobbyists live in the shadows of such giants. The amazing thing, would be to get some ancient networking protocol to work AT ALL. I suggest humility, and avoid the simple notion "I can plug it in - so it will work, or could".
Time-out, please. I appreciate there are mega-minds out there, that talk TCP/IP and know protocols forwards and backwards. They can manipulate Linux like Play-Doh. That's not me, I'm an old person that knows 8-bit computers from original use. I have those skills and others, I have some training, I know what "skills" means. And so my situation, likely applies to a number of other people, who own other 8-bit or 16-bit computers, who post in or read this list. We aren't mega-minds, but we know some stuff; and we have some vintage stuff. Here's the deal as I see it. There's dead networking protocols from the 8-bit and 16-bit microcomputing era. ARCnet. NOvell. Token Ring. And not quite dead in the Mac world, Appletalk (over Phonenet or on Ethernet). These are not in (much) use today. Hardware and software for them are scattered to the winds. Current hardware is powerful - but these are dead technologies, the two don't mix well. Maybe there's some Web page or email discussion about each of these - but it's likely to be a group of mega-minds who will be talking about packet-drivers, who worked for Novell or Tandy, who are writing them in TK-TCL, to run on Sparc stations - completely above my pay grade, out of my reach, beyond my experiences and most of my vintage hardware. OK? And not just me. The issue isn't "let's make protocol converters so ARCnet can talk TCP/IP and we can print pages and do Web browsers on TRS-80s". The issue is "look - I have ARCnet running between two computers!" OK? Base hits, not home-runs with bases loaded and fireworks. Sorry to be boring. But - I'm not gonna be able to gather, say, ARCnet hardware and software together on some S-100 boxes and *make them work by myself*. OK? But I doubt someone else is going to stop what THEY are doing, ask me to do nothing otherwise too, work by email from hundreds or thousand of miles away - all to try to get some 4 MHz Z80 S-100 box to talk to an ARCnet card to some hub to some other ARCnet node (which I don't have or would be yet another S-100 box). Same with a pair of IBM-PC's or a pair of Tandy boxes, etc. That's the general circumstance. People with some vintage computers, some vintage networking hardware - but no means to access a working test networking setup, no skills-set to reconstruct such a setup from scratch. OK? But *maybe* if there's a group of smart-enough people with hardware-in-hand, they can come up with such a working test setup, for ONE networking protocol, that can be replicated by mere mortals. *Maybe* they can exhibit such a setup, or bring it to some workshops. *Maybe* us mere mortals can bring our toys into that situation, and try to make them work. To do so efficiently, we would have to do some homework before-hand, right? To have the hardware ready, to have some software ready, some kind of testing, some kind of performance established - before we show up. Right? Think that through. What's needed to make that happen? It means those "smart people" provide some resources online, so they can be read and downloaded, so these 8-bit owners can read-up, prepare hardware, find resources, etc. OK? I'm just walking backwards, through some planning. And *that's the hard part*, for the (excuse me) mega-minds. It's fun to talk about complicated things, among other people who already know that stuff. It's less fun, to explain it to those who don't. And the *least fun*? Laying out the ground-work, Web pages, drivers, software, docs on hardware (even finding hardware) - so we of lesser mind and resource, can run this stuff ourselves, learn how to make it run ourselves or with a little help. *That's* the situation as I see it. That's my view, from my experiences. Now - if a few people who know this stuff, just want to set up some Novell or other dead-network and show it off - that's fine. It would be interesting, and certainly a challenge, and is a worthy exhibit. My memory is short - maybe such networking exhibits have occurred at VCF-East in the past. But how about, *preserving* that knowledge and "passing it forward"? How do you do *that*? That's what I'm talking about. That's what happens, afterwards. For all I know, others HAVE done this - and have a Web site, and it's all there. Haven't checked lately, pardon my ignorance. And if that's just outside the scope of what Bill Degnan and some other folks are talking about, of course that's entirely up to them to not do. This is a hobby and people do what pleases them and interests them. I'm presenting a point of view about Bill Degnan's proposal; I'm responding to his "idea for an exhibit". These are my responses ... and I'm done. Herb Johnson -- Herbert R. Johnson, New Jersey in the USA http://www.retrotechnology.com OR .net preserve, recover, restore 1970's computing email: hjohnson AT retrotechnology DOT com or try later herbjohnson AT retrotechnology DOT info
On Sat, May 26, 2018 at 11:15 AM, Herb Johnson via vcf-midatlantic < vcf-midatlantic@lists.vintagecomputerfederation.org> wrote:
Time-out, please.
I appreciate there are mega-minds out there, that talk TCP/IP and know protocols forwards and backwards. They can manipulate Linux like Play-Doh. That's not me, I'm an old person that knows 8-bit computers from original use. I have those skills and others, I have some training, I know what "skills" means. And so my situation, likely applies to a number of other people, who own other 8-bit or 16-bit computers, who post in or read this list. We aren't mega-minds, but we know some stuff; and we have some vintage stuff.
Here's the deal as I see it. There's dead networking protocols from the 8-bit and 16-bit microcomputing era. ARCnet. NOvell. Token Ring. And not quite dead in the Mac world, Appletalk (over Phonenet or on Ethernet). These are not in (much) use today. Hardware and software for them are scattered to the winds. Current hardware is powerful - but these are dead technologies, the two don't mix well.
Maybe there's some Web page or email discussion about each of these - but it's likely to be a group of mega-minds who will be talking about packet-drivers, who worked for Novell or Tandy, who are writing them in TK-TCL, to run on Sparc stations - completely above my pay grade, out of my reach, beyond my experiences and most of my vintage hardware. OK? And not just me.
The issue isn't "let's make protocol converters so ARCnet can talk TCP/IP and we can print pages and do Web browsers on TRS-80s". The issue is "look - I have ARCnet running between two computers!" OK? Base hits, not home-runs with bases loaded and fireworks. Sorry to be boring.
<anip>
*That's* the situation as I see it. That's my view, from my experiences.
Now - if a few people who know this stuff, just want to set up some Novell or other dead-network and show it off - that's fine. It would be interesting, and certainly a challenge, and is a worthy exhibit. My memory is short - maybe such networking exhibits have occurred at VCF-East in the past. But how about, *preserving* that knowledge and "passing it forward"? How do you do *that*? That's what I'm talking about. That's what happens, afterwards.
For all I know, others HAVE done this - and have a Web site, and it's all there. Haven't checked lately, pardon my ignorance.
And if that's just outside the scope of what Bill Degnan and some other folks are talking about, of course that's entirely up to them to not do. This is a hobby and people do what pleases them and interests them. I'm presenting a point of view about Bill Degnan's proposal; I'm responding to his "idea for an exhibit". These are my responses ... and I'm done.
Herb Johnson
From a exhibit perspective I'd rather see an actual Novell or same era exhibit than an exhibit that attempts to connect an old protocol to the modern world, something that did not exist then. Isn't as educational or on-topic to our group. My intention was to teach/demo what was then as it was then.
Bill
On 05/26/2018 01:48 PM, Bill Degnan via vcf-midatlantic wrote:
From a exhibit perspective I'd rather see an actual Novell or same era exhibit than an exhibit that attempts to connect an old protocol to the modern world, something that did not exist then. Isn't as educational or on-topic to our group. My intention was to teach/demo what was then as it was then.
Cool, that helps. But may I ask when was 'then'? I would think that mostly you'll be working with IBM PCs. I looked up Netware and it was released in 1983 and the Wikipedia notes that there was support for PCs (DOS) and CP/M. I've not seen anything but DOS. I learned Netware in 1992 and I had customers that still had Netware until 1998 (most still on 3.x and not 4.x). Has anyone seen a CP/M machine with Netware? -- Linux Home Automation Neil Cherry ncherry@linuxha.com http://www.linuxha.com/ Main site http://linuxha.blogspot.com/ My HA Blog Author of: Linux Smart Homes For Dummies
On 05/26/2018 11:15 AM, Herb Johnson via vcf-midatlantic wrote:
Time-out, please.
I appreciate there are mega-minds out there, that talk TCP/IP and know protocols forwards and backwards. They can manipulate Linux like Play-Doh. That's not me, I'm an old person that knows 8-bit computers from original use. I have those skills and others, I have some training, I know what "skills" means. And so my situation, likely applies to a number of other people, who own other 8-bit or 16-bit computers, who post in or read this list. We aren't mega-minds, but we know some stuff; and we have some vintage stuff.
Guilty (except for the Mega Mind - though going from super villain to super hero might be pretty cool. yes a movie reference ;-) ). Also, you are a mega-mind, just not on this subject. Bill's email clarifies the networking to Novel. I may have an NE2000 board and maybe a clone board he can use. I don't want to take my 386SX and convert it's drive from Linux (1.2) to DOS or Windows so I don't have a PC for him. -- Linux Home Automation Neil Cherry ncherry@linuxha.com http://www.linuxha.com/ Main site http://linuxha.blogspot.com/ My HA Blog Author of: Linux Smart Homes For Dummies
On Sun, May 27, 2018 at 10:18 AM, Neil Cherry via vcf-midatlantic < vcf-midatlantic@lists.vintagecomputerfederation.org> wrote:
On 05/26/2018 11:15 AM, Herb Johnson via vcf-midatlantic wrote:
Time-out, please.
I appreciate there are mega-minds out there, that talk TCP/IP and know protocols forwards and backwards. They can manipulate Linux like Play-Doh. That's not me, I'm an old person that knows 8-bit computers from original use. I have those skills and others, I have some training, I know what "skills" means. And so my situation, likely applies to a number of other people, who own other 8-bit or 16-bit computers, who post in or read this list. We aren't mega-minds, but we know some stuff; and we have some vintage stuff.
Guilty (except for the Mega Mind - though going from super villain to super hero might be pretty cool. yes a movie reference ;-) ).
Also, you are a mega-mind, just not on this subject. Bill's email clarifies the networking to Novel. I may have an NE2000 board and maybe a clone board he can use. I don't want to take my 386SX and convert it's drive from Linux (1.2) to DOS or Windows so I don't have a PC for him.
Oh, I don't want to make a Novell exhibit, I was suggesting someone who supported Novell in it's heyday or wishes to learn it do the exhibit because we need something new and we need to broaden the scope of the types of exhibits we present. 1989 would be 30 years ago, I'd avoid something too new because we are still generally-speaking an 80's and older-scope event. The exhibitor should try to create a Novell office environment. An office department that does CAD drawings and processing for example, wordpressing, dbase III, custom C menu interfaces, etc. b
On 05/27/2018 10:36 AM, Bill Degnan via vcf-midatlantic wrote:
Oh, I don't want to make a Novell exhibit, I was suggesting someone who supported Novell in it's heyday or wishes to learn it do the exhibit because we need something new and we need to broaden the scope of the types of exhibits we present. 1989 would be 30 years ago, I'd avoid something too new because we are still generally-speaking an 80's and older-scope event.
The exhibitor should try to create a Novell office environment. An office department that does CAD drawings and processing for example, wordpressing, dbase III, custom C menu interfaces, etc.
Sorry about that Bill. My offer to borrow my NE2000 still stand to the gentleman who was planning the network exhibit, if he wants it. Now I'm going to shoot myself in the foot and say perhaps a counter exhibit of an industrial communications network would be in order. I have a Circuit Cellar HCS II (Home Control System) which is based on Steve Ciarcia's BCC180 (the industrial version). It has a multi-drop network (R485, half-duplex) and is all 8-bit. I'll have to see how hard this will be to resurrect a complete system. I have the compete information on this system. :-) Does anyone else have information of networking on the 8 bit computers? -- Linux Home Automation Neil Cherry ncherry@linuxha.com http://www.linuxha.com/ Main site http://linuxha.blogspot.com/ My HA Blog Author of: Linux Smart Homes For Dummies
Herb Johnson wrote:
A few S-100 companies supported ARCnet. Boards were available from Compupro and a few others companies. I've collected a few ARCnet devices and S-100 cards. But this stuff is outside my digital EE experience. And if I got one of these cards running, what would it talk to?
One of the reasons I suggested LanTastic. As I recall, the low level protocol was simple enough to implement on S-100. Maybe requiring a 4MHz z80? I would be cool to have multiple exhibits all networked together. Bill S. --- This email has been checked for viruses by Avast antivirus software. https://www.avast.com/antivirus
participants (8)
-
Bill Degnan -
Dave McGuire -
David Riley -
Ethan O'Toole -
Herb Johnson -
Neil Cherry -
W2HX -
William Sudbrink