Ethernet at 40: Its daddy reveals its turbulent youth

When Bob Metcalfe, the prime mover behind the invention of Ethernet, recently visited the site of that invention, Xerox's Palo Alto Research Center (PARC), The Reg had the opportunity to sit down with him to discuss the history of Ethernet, its advantages over Token Ring, and IBM's perfidy. Metcalfe was in town to promote a …

[William Boyle]

Robert - get back into real engineering!

I have known RM since the early 1980's - I sold 3com their first 100 PC's... Later, when I was a principal engineer at a major Boston area software company, we used to meet for dinner before IEEE meetings in the area (usually at Mitre). Now, I am a senior systems engineer at a tier-one mobile phone manufacturer - Robert and I are about the same age. So, I can only say ... "Robert! It is time you get back to your roots!".

P.S. I have utmost respect for RM. He has been one of my tech heroes for 30 years now. :-)

[Long John Brass]

Old ARCnet jocky here

I kinda miss ARCnet sometimes

Most of the benefits of token-ring and most of the benefits of early Ethernet

Still remember soldering up 4-way passive hubs *sniff*

[Fred Flintstone]

Re: Old ARCnet jocky here

Yup - I remember soldering terminators because the "official" ones were ridiculously overpriced. Provided people kept their fingers off the cabling the stuff was actually quite stable, and the cards were pretty reasonably priced too (except the 3COM one we stuck in the server). However, try to set up a WAN and it got very expensive..

God knowns how many miles of coax I've used in those days but it was a *lot* :).

[Christian Berger]

ARCnet still lives

It's used, for example in broadcast engineering for real time control. There are also rumors of new industrial installations using it. So it's still around in some new installations.

[Joe H.]

3 Com and early Ethernet

Once upon a time 3 Com shipped a customer a whole bunch of 10 meg ISA BUS Ethernet network cards, about 200 if I remember correctly, and these were fitted into shiny new 486 processor boxen and duly connected to 10 meg Ethernet hubs and switches for their data processing folks who were moving into a nice new office.

When it came time to test, nothing seemed to work correctly and even the tier 3 engineers were stumped by the fact that pings and trace routes would go nowhere. Much finger pointing and recriminations ensued until someone brought in a Network General sniffer and did some packet captures.

It turned out that all of the 3 Com network cards had been accidentally burned with the same mac address and that mac address was the only one that showed up in the packet traces.

I was a remote technician on maintenance support for the hubs and switches and was just as amused as everyone on our team that the problem was not ours, good times.

[Anonymous Coward]

Re: 3 Com and early Ethernet

Pretty sure one of the vendors did it again later with PCI cards , maybe even 3com..possibly the 3905CX (yup didn't even look that up). it was a bloody nightmare.

[Anonymous Coward]

Wasted time

As a freshly graduated pup in the late 70's, I worked at a British university computer dept doing research into token ring fault tolerance. Our particular ivory tower had strong links with Cambridge University Computer Laboratory and its eponymous Ring (I once had the privilege of meeting Andy Hopper and Maurice Wilkes - yay!). Of course, that meant I was indoctrinated with all of the "ethernet bad, ring good" propaganda. After finally escaping to a proper job, I had to set up a network that functioned for real and ended up with ethernet. I still can't believe all that wasted time fscking about with s0dding rings.

Don't get me wrong, the ring was very clever and put together by some extraordinarily smart people, but ethernet was better in most respects.

(Ok, maybe it wasn't all wasted time, if one considers the valuable education)

Cue downvotes for nostalgic lovers of The Ring...

[Matt Bryant]

Re: Wasted time

"......Don't get me wrong, the ring was very clever and put together by some extraordinarily smart people, but ethernet was better in most respects....." Agreed. I was introduced to Ring, loved it, it just worked, but it was very sad to see the lengths to which the ant-Ethernet crowd went to. Ethernet simply was better, but the technical discussion got hijacked by business people that saw Ethernet as a threat to their incomes, and then they put monetary pressure on technical and academic people to join one camp or the other. A lesson for those that like to cast M$ as the Big Bad in the PC wars - they were just following the business practices set by IBM.

[KA1AXY]

Re: Wasted time

As a former token ring network developer, I could not ahree with you more. Glad to see the end of it, especially 16 meg TR. What a piece of proprietary junk - at 3x the cost of 100 meg ethernet!

[ThomH]

Ethernet is the Ayn Rand network?

There's no central authority instructing the nodes to act; they discover whether it's safe to broadcast through their own local observation. The lack of a centralised actor and the ostensible resulting chaos leads to a more efficient overall system.

I'm not a libertarian but I can see there's a reasonable argument in there.

[Anonymous Coward]

Re: Ethernet is the Ayn Rand network?

Ayn Rand was a neo-fascist - what could be more centrist than that ...

[Destroy All Monsters]

Re: Ethernet is the Ayn Rand network?

Ayn Rand was a neo-fascist

Downvoted from the incursion from the left-winger's pigpen.

Ayn Rand was not a good writer and crazy radical, but had all the normative ethics written down for analysis. She detested socialist leechers and failures, in particular the socialist failures resorting to violence, so how can she be "neo-fascist"? What does that even mean?

Hold on, are you trolling?

[PyLETS]

Re: Ethernet is the Ayn Rand network?

Ayn Rand probably wasn't a fascist, but she was certainly a hypocrite for trying to create a moral justification for why her followers had to do what she told them, when she had taught them that morality was obsolete.

[Peter Murphy]

Re: Ethernet is the Ayn Rand network?

It's not so easy to draw a political moral from the development of the Ethernet. If it's libertarian, it's more social libertarian ala Noam Chomsky rather than the capitalist libertopia of Ayn Rand. The nodes are just communicating with each other, after all. They're sending out packets, not accummulating them.

The only connection to Atlas Strugged is this. IBM tried to go Galt. They failed.

[Christian Berger]

That's not her philosophy

Rand wants everyone to be selfish and only follow their own goals because she doesn't understand that there is a difference between individual goals and the common good. Ethernet is actually a good example why she is wrong.

Ethernet works, because there are strict rules everyone must adhere to. If there is a collision you have to wait a random time. With each collision the timespan of that random number will increase. This rule throttles down stations when the network is full, to share the bandwidth more or less fairly between them.

Now an Any Rand network card would simply ignore the standard and re-transmission the packet right away. That way it will always give the card 100% of the network if it wishes to. However if more people act like this, there will _only_ be collisions. Once there is a collision, both stations will try again at the same time, they will do so until one of them gives up. Instead of sharing the bandwidth fairly between the stations, those few stations ruin the network for everyone.

[Daniel B.]

Re: That's not her philosophy

Oh the irony. Metcalfe made it big by doing a network that runs *against* Ayn Rand ideals. Interesting!

[Ian Ringrose]

Re: Ethernet is the Ayn Rand network?

In real life Ethernet worked very well, most networks had a few file servers that sent out most of the network packets, clients did not request the next bit of a file until they have got a response to the last request.

So yes, Ethernet has no central authority, but in real life it was very common for 1% of the machines on a network to be sending out 80% of the packets. DEC published a research report on “real life performance of Ethernet” that showed how much better Ethernet worked then you would expect if you modelled it with random unrelated packets.

What we call “Ethernet today” is nothing like what Ethernet was like, as today it is all switched, in the old days all the computers connected to the same cable using T-Connectors; there was no electronics between the computers, just plain cable.

[Roland6]

Re: What we call “Ethernet today” is nothing like what Ethernet was like

Yes, I noted how this aspect of the evolution of 'Ethernet' was totally glossed over.

The reason why 10Base-T LAN is called Ethernet is largely down to marketing and lobbying of the IEEE by interested parties.

There were debates within the IEEE (and contributing organisations) as to whether 10Base-T really was an extension to 802.3 or should be a new 802.n standard. What seemed to finally swing the case for 10Base-T to be an extension of 802.3 and hence be called 'Ethernet' was that 'Ethernet' had a good market presence and image and hence vendors would find it easier to gain market acceptance of 10Base-T solutions if they were branded "Ethernet", along with the weight of agreement among the contributing members. Once the decision was made, the rest was history...

I suggest that 802.11/Wi-Fi is the only other 802 standard to have gain a similar level of market awareness.

[Anonymous Coward]

U.S. Patent #4,063,220 - Ethernet Patent ..

"Multipoint data communication system with collision detection U.S. Patent #4,063,220

[Anonymous Coward]

IBM's arrogance was that they could make standards ..

I recall reading a story where IBM produced a printer, the standard one and the advanced (more expensive) one, where the only difference being moving a belt on a drive wheel to speed up the printing.

[Anonymous Coward]

Re: IBM's arrogance was that they could make standards ..

Everyone does this. You really think the difference between a low end smartphone and a high end smartphone is hundreds of dollars?

I remember having an Amstrad Fidelity satellite receiver which didn't work. Someone sold me a similar model that did work but didn't have the remote control. I opened both of them and found that all I needed to do was swap the control boards over. The cost difference was nothing, but there was probably a £50 difference in price.

[jason 7]

Re: IBM's arrogance was that they could make standards ..

Yeah its like digital cameras. The £70 one and the £400 have the same chips and sensors its just that the £70 one has the settings reduced in the firmware to make the £400 one look far better.

[Stu_The_Jock]

Re: IBM's arrogance was that they could make standards ..

Actually I think you'll find one of the differences between £70 and £400 cameras is also in the quality of the lens and zoom mechanism, the quality (not the pixel count) of the sensor to allow lower light shooting, the intelligence to drive the auto-focus at a decent speed, etc etc etc. A camera, like any techy device contains many separate parts that can influence the final product. No point having the worlds best sensor and autofocus systems and then throwing in a cheap plastic lens system.

[Dave Ingram]

Re: IBM's arrogance was that they could make standards ..

Same chips and sensors maybe, but did you stop to consider that the lenses could be rather different?

A gazillion megapickels is wasted if the optics are not good enough. That's why a 6Mp photo taken with a Minolta Dynax 5D can be printed as an 8"x12" and look nice, but an 8Mp smart(arse) phone photo is barely passable as a 6"x4" print.

[Anonymous Coward]

Re: IBM's arrogance was that they could make standards ..

@Dave Ingram; in that case, it's also because the sensors *will* be different- same amount of pixels, but the one in the smartphone is likely to be smaller (and cheaper) with less light-gathering capacity per pixel, and hence more noise and more noise-reduction required compared to the one in a typical DSLR.

[Titus Aduxass]

Re: IBM's arrogance was that they could make standards ..

Clearly you've never heard of lenses.

[Anonymous Coward]

Re: IBM's arrogance was that they could make standards ..

Some mainframe products from various manufacturers had optional extras that merely needed a single link on the backplane to be wire-wrapped to activate them. Intermediate mainframe models had added hardware delays to slow them down. It could even be a case of just loading a different firmware to effect a performance upgrade. They were all ways to standardise the manufacture - and the customer paid for a particular performance.

[Wensleydale Cheese]

Re: IBM's arrogance was that they could make standards ..

I also recall printers and disk drives being "field upgradeable", and that was achieved by snipping a link to get increased speed and capacity respectively.

[peter 45]

Re: IBM's arrogance was that they could make standards ..

I setup a little business at Uni buying single sided FDDs and selling them on as double sided. All you had to do was open it up and reconnect a disconnected cable. My profit was half the difference between the priced the company sold the two versions.

Icon for the beer the little sideline kept buying for me.

[Nuke]

@dgharmon - Re: IBM's arrogance was that they could make standards ..

Wrote :- "IBM produced a printer .. standard .. and advanced (more expensive) ... the only difference being moving a belt on a drive wheel to speed up the printing."

Anoher example was Windows NT Desktop vs the far more expensive NT server. The CD contained exactly the same code, except for a flag somewhere in the installation script that allowed the server features to be installed or not.

[Mayhem]

Re: @dgharmon - IBM's arrogance was that they could make standards ..

It is also fairly common to have a lower quality product when the production lines are getting up to speed, especially for things like graphics cards.

The ones that pass QA get sold as the top end product. The ones that fail QA - but still work - get sold as the varying grades of cheaper product.

The ones that fail QA and don't work get sold for OEM built in cards :p

[Herby]

What? No Picture?

Of the beloved napkin?

Please add it (I'm sure it is somewhere out there!).

[Marvin the Martian]

What is even more missing: an alternate history.

If (as he says should maybe have happened) Simonyi had written Ethernet instead, we'd have a well-functioning open standards wysiwyg text editor, and an expensive (even more than TR) proprietary network interface that has Clippy the Paperclip trying to talk us into doing stuff?

/The good half of that outcome is more or less there (hello, LibreOffice4.0).

[JimC]

twisted Pair...

I know almost everyone says how wonderful twisted pair cabling was, but one of the major advantages of ethernet for us over Token ring was that it didn't require star wiring for every damn device. By and large we weren't installing in purpose built office buildings but in older, often historic buildings with no cable space. Getting permission for a run of thinnet, maybe with a thicknet backbone, or even earlier corvus omninet which was a STP bus, was straightforward, getting permission for vast wodges of twisted pair for TR would have been next to impossible. And then anytime anyone wanted a new device it would have been get the cablers in. Mind you that hasn't changed, because every time we think we've flood wired an office they have a reorg and cram yet more tinier desks in at an even higher density...

A few years on of course the managers all had their own devices on their desk and those pig ugly runs of trunking suddenly weren't a problem any more... Especially if it meant no more engineers crawling round our office with a TDR trying to work out where the break actually was...

[Christian Berger]

Re: twisted Pair...

Wait, you had a TDR? You lucky bastard :) Most people had to work with multimeters. You'd open the Ether and check what direction the fault was.

[Neil Barnes]

Re: twisted Pair...

But at least with Ether you didn't have to crawl around the floor later sweeping up the tokens that had fallen out...

[Yet Another Anonymous coward]

Re: twisted Pair...

But at least you could re-use and recycle the tokens. With aethernet everything is just dumped into the aether - polluting it

[JimC]

Re: twisted Pair...

To be quite honest the TDR wasn't nearly as useful as you'd think it ought to be. Because you never knew how many lengths of thin net had been chained together and wrapped round various desks a distance to break wasn't that useful. It was often just as quick to just run through the offices with a spare terminator.

Best thing to have was when Jane and I were throughly debugging out a rather chaotic install in the Fire Service HQ - a Victorian mansion on about 7 diffeent levels - and we managed to borrow a couple of walkie/talkie radios. "Ok, which room are you in now" "Terminator off NOW"...

I bet I've still got an animal made of thin net connectors somewhere [grin]

[meanioni]

Re: twisted Pair...

Ugh star wiring - surprised no-one mentioned it - for me the main killer of Token Ring.... I used to work in the insurance company Commercial Union who lived, ate and breathed IBM. They backed Token ring over Ethernet, OS2 over Windows, Smartsuite over Office, Lotus Notes over... well any other half decent email programme. I set up my own little enclave with a peer to peer Windows for Worksgroups network (remember that!) running over Ethernet, using Office.... I fought off battles from my IT Department...

Then about 2 years later the nerds in the IT centre panicked and came to us as the most experienced people using these technologies.... ah those were the days :-)

[PT]

Re: twisted Pair...

"...a peer to peer Windows for Worksgroups network (remember that!)"

Yes - I'll never forget the blinding epiphany when I discovered I could add computers on the existing network without having to pay the Novell tax.

[Nuke]

@meanioni - Re: twisted Pair...

Wrote :- "I set up my own little enclave with a peer to peer Windows for Worksgroups network (remember that!) running over Ethernet, using Office.... "

Windows for Workgroups?! Nostalgia for that that PoS? Please tell me you're kidding.

In fact I had WfW on an old PC until just 6 months ago - I used it with an old but very fast scanner that had no drivers for anything later.

[meanioni]

Re: @meanioni - twisted Pair...

My post was tongue in cheek, so yes I was kidding!

WfW is in the same zone as Win 95, Win ME (shudder), Vista, etc....

[Ian Ringrose]

Re: twisted Pair...

At least by the time Ethernet supported twisted pairs, you could support additional devices on a single pair just by using a very CHEAP “repeater”. So it was more a star of stars of stars setup for Ethernet.

You could get devices to do the same for Token Rings however they were VERY expensive

[KA1AXY]

Re: twisted Pair...

Just you try getting a 16 meg unshielded twisted pair token ring switch through EMI testing...not fun at all.

[Anonymous Coward]

Ah Ethernet

The joy of accidental linkages to dirty earth from exposed connector metal; threading a long AUI cable through a false floor - then discovering it was the wrong way round. Bee-stings that vibrated when a jet took off nearby. Badly engaged AUI slide-locks causing connectors to hang off - so they transmitted but didn't receive.

The thin coax was prone to user interference: kinked cables; disconnected cables; extra terminators; different impedance cable sections; departments installing "private" bridges that put a loop into the Company Wan. User software - usually a cloned install or config file - that used the same private MAC address to override the hardware one.

Some units had a deviant clocking rate that other suppliers' units couldn't sync with. Excessive dribble bits also caused compatibility problems. Some suppliers modified the back-off algorithm to give their kit unwarranted priority on a collision - so their performance measurements under high loads were better than the competition. Too many people thought that any "collisions" were indicative of a network fault - it should have been called "contention".

Initially the 10mbps technology at the physical layer was cutting edge, and expensive, before standardised Manchester encoding components made it easier. That possibly slowed down the initial market penetration.

[Don H]

Performance

The cabling advantages of 10base-2 went away with 10base-T, which went back to a star topology. It was educational to watch the utilization lights versus the collision lights on the hubs, behaving exactly as the theory predicted. Also, most original T-R cabling was done with Type 1, which was STP. It worked very well with Ethernet, either with baluns to terminate 10base-2 or adapters to terminate RJ-45. Installations going from T-R to 10base-T didn't have to replace their building wiring.

Back in 1991 I built a pair of lines for duplicating disks. Each line consisted of a server PC and 10 clients which pulled files from the server. One line was linked by 4Mbps T-R, one by 10Mbps 10base-2. For 1-to-1 duplicating, the T-R line took 12 minutes and the Ethernet line took 10 minutes. For 1-to-10 duplicating, the T-R line took 20 minutes and the Ethernet line took an hour and 20.

Eventually, switches replaced hubs and T-R's main performance advantage evaporated.

[P. Lee]

Re: Performance

I'm not sure about Ethernet being better in the early days. Cheaper and simpler, perhaps, and beaconing was a bad thing, but troubleshooting non-star wiring breaks was horrible. Before switches, collisions were a problem which meant the headline Ethernet speed meant little on a congested network.

The joy of Ethernet was the ad hoc doom party - disconnecting from the office lan was easy, or at home nothing more than a bit of thinnet was needed.

[Destroy All Monsters]

Re: Performance

> The joy of Ethernet was the ad hoc doom party

Oh no, memories!

[Anonymous Coward]

10 Base T

The class idiot at college who would steal a terminator or disconnect one of the computers thus taking down the entire network. Everyone would then have to hunt to see where the break in the network was.

+ idiot system admin that insisted Windows 3 was loaded from a network drive. So each computer would take about 5 minutes to load windows on a bad day.

Glad to see the back of that system!

[Fred Flintstone]

Re: 10 Base T

Actually, I didn't find Worries for Workgroups that much of an improvement (I've called it that practically from day 1).

On the plus side, it prepared me to jump to Linux based SMB file sharing when that showed up later, so maybe it wasn't all bad.. :)

[Daniel B.]

Worries for Workgroups

Ah, that thing. Windows 3.11 with DOS 6.22, which for some weird reason broke compatibility with all the stuff in windows 3.1 and DOS 6.2. I kept the 3.1/6.2 combo because it wasn't worth doing the upgrade.

[Anonymous Coward]

Re: The Classic Idiot...

Sales rep calls from the demo room... Hugely important meet, net gone down, just as he was telling the prospective customers how good it was.

Yes, we were wondering why, too! Turned out he'd disconnected the cable from the back of the demo-room machine to show them.

An installation I did... Customer had done their own cabling. Connected all my new gear up... No network. A minute's thought, and I realised that, as I'd been attaching machines, I had not seen a single terminator, anywhere. Yes... He thought that thin ethernet should be an unbroken ring! Luckily I always had a tool box full of useful things.

[jason 7]

Wouldnt mind.....

...a little bit of Token Ring's efficiency creeping into Ethernet.

Always feel with Ethernet I have a huge V8 Muscle car that's always stuck in first gear.

[Christian Berger]

Re: Wouldnt mind.....

Well you have that right now. There are switches now on Ethernet. The issue everyone had with Ethernet is gone, there are no collisions any more. What you have now are fairly laminar streams of data.

[banjomike]

Ring was too expensive

£200+ for one Token Ring card, £800 for a MAU. Geez. I remember rewiring a Token-ring office with ethernet for less than the price of ONE additional Token Ring card.

[Pet Peeve]

Re: Ring was too expensive

Even the cabling was expensive with token ring. Our office used it for a few years, and the cables were as fat as your index finger and had a huge clip-on connector on the end. Serious grumbles if a pin broke because someone moved their PC with the connector attached.

I seem to remember that it wasn't strictly star topology - that you could do something to daisy chain devices together and then back to the network run, but it's been almost 20 years since I had daily exposure to that stuff - don't remember the details.

[Tim99]

Re: Ring was too expensive

@Pet Peeve.

You could connect MAUs together by using main link connectors...

[Daniel B.]

Re: Ring was too expensive

IIRC Token Ring was actually *RING* topology, hence the name.

The other ring topology network I can remember is LocalTalk, which for years was what our home network had.

[Peter Simpson 1]

Re: Ring was too expensive

IIRC, the Token Ring cost included a license...for the MAC code, whereas Ethernet MAC code was so simple it didn't need a license. TR MACs needed all kinds of code to manage token forwarding, master election and such. Ethernet "just worked", although it worked a lot better when we got away from coax and onto twisted pair...

[TeeCee]

Re: Ring was too expensive

Er, rubbish. A MAC code is a MAC code, all it has to be is unique. You could assign your own with TR, if you really had nothing better to do with your life, so what they actually were was irrelevant.

The cost was in the intelligence of the cards. Every token-ring node can act as a master (first one on the ring gets it until such time as something you've designated as master turns up and takes it) and also performs diagnostic monitoring on the upstream device, generating alerts if necessary.

Also, the TR protocol has the capability to route between LANs built in, which adds to the complexity. Once a route is established, the end nodes are expected to address that route correctly themselves. The source-routing bridges only supply transport and discovery services.

Ethernet cards were dumb and much cheaper.

Upsides: TR can always tell you exactly where a fault has occurred and usually exactly what the fault is. They'll also tell you exactly what's where and when it appeared there. Finally, adding rings to a network is trivial as it works it all out itself when a new bridge shows up.

Downside: cost.

[banjomike]

Re: Ring was too expensive

I bought some Attachmate Token Ring cards when working at Suna Alliance. Two of those cards had identical MAC codes. You could almost hear the beaconing out in the car park. With two identical MACs the network monitor could not identify where the error was. Chaos.

[Anonymous Coward]

Re: Ring was too expensive

Like many of the comments here time has dulled the memory.

"the cables were as fat as your index finger and had a huge clip-on connector on the end"

That sounds like thick ethernet. One fairly common approach was to run the vertical backbone on thick ethernet but distribute on each floor using token ring. The site I saw where the network was a real support nightmare they did it the other way round, token ring backbone, ethernet on the floor - delivering the worst of both.

Other comments here are comparing token ring with ethernet AFTER the time when ethernet was seriously expensive and a right PITA to implement and support. Ethernet got much faster, much cheaper and more resilient, token ring only got a bit faster and the cost didn't shift so it ended up relatively expensive.

[TeeCee]

Re: Ring was too expensive

Nope, that's good old IBM Type 1, or possibly Type 6, with those unisex plugs on each end. Even the wallports were just bits of plastic that accepted a unisex plug in the back.

UTP came somewhat later to token-ring. Originally everything hooked up using those connectors. Making 'em up was a bitch too. In theory, as they snapped together it should have been simple. No crimping, screwing or any other "ing"s required. In practice, getting all the bits in the right place and keeping them there while snapping the casing together was like juggling six balls while sat on a unicycle.

[Sir Crispalot]

IBM NHBR

I joined IBM in 2000 and they were still using token ring at North Harbour. It wasn't until a year or two later that the whole site was converted to ethernet. Pretty shocking!

[Jock in a Frock]

Ah, sunny Cosham.

I work for a telco who supplies services to IBM North Harbour. Luckily I only worked onyour WAN links, so never had to work with Thicknet or TR. I do remember the banks & banks of patch panels there though. I thought they were mains cables the fist time I saw them.

[John Smith 19]

Ethernet Vs Ring Practical Vs Theoretical.

Rings have 2 path failures to isolate a node and calculable maximum latency. Handy if you're implementing an aircraft flight control system in a high noise environment.

But IRL most nodes did not want to talk all the time and traffic was quite "bursty" (like the print job to a printer) and the EMI resistance did not need to be important.

Like Betamax Vs VHS?

Except current Ethernets are 100x better than the original.

[Johan Bastiaansen]

What? No bashing...

on account of the Ayn Rand comments? Is this the same group that was gleeing over that born-again idiot quitting his job yesterday?

Not so brave when someone rich a mighty has a similar mental handicap eh.

So when your boss want's a vanity number on his wage slip, you would be bending over backwards?

[Pet Peeve]

Re: What? No bashing...

It's troll bait, and it looks like we caught one.

[Fred Flintstone]

Re: What? No bashing...

What? No bashing on account of the Ayn Rand comments?

It makes this discussion FCC rated.

Must be able to sensibly accept interference, should not cause interference. Yup. Works for me.

[Anonymous Coward]

Re: What? No bashing...

Metcalfe complains quite a bit about IBM's FUD, but he himself in his InfoWorld columns during the '90s made it a point to persistently refer to "open source" as "open sores", and championed the work of libertarian economists who claimed it was not viable as a development or economic model. Some 20 years later, I would say he and those economists have been proven wrong, and by no small measure.

Open source succeeded for much the same reasons Ethernet did, but Metcalfe let his own ideological fundamentalism blind him to its potential and opportunities. And to place Ayn Rand, a writer of ideological fiction, ahead of the great Anglo-Scots secular philosophers like Smith, Locke, Mill, and Hume, is typical of the American tendency to confuse entertainment with actual thought.

[cmrayer]

Errors and omissions

12Mb Token Ring, er 16Mb I think! Also Early Token Release fixed the lost token issue but really it was all down to cost once the PC CPU had enough spare cycles to do in software what the TR NIC did in hardware. Funny there was no mention of ATM, that was a great technology with LAN/WAN QoS/COS all built in at layer 2 and the chips were much more like an Ethernet chip in production costs. Also there was no mention of Madge networks a British company who at one time were nearly as big as 3Com and had over 80% of the Token Ring market. They also pioneered switched TR and 100Mb TR in the early 1990s. Finally Type 1 is alive and well in some of our older buildings running 100Mb FDX Ethernet and the rats can't chew through it ;-)

[brianpuente]

Re: Errors and omissions

I worked for Madge from 1997 to 2000. We had the best technology if you ask me. I did tech support for the token ring cards, and switches, at their American office, in San Jose California. It was sad to see the switch to Ethernet kill our company slowly each quarter. Even more sad was when we bought our only non-IBM competitor, Olicom in 1999. They were our arch enemies, but at least we had enemies. Shortly after that, I was laid off, and ironically went to work for 3Com in Santa Clara. 3Com made some pretty stupid moves let me tell you (Audrey anyone?? This good for nothing, over priced piece of crap was given away to everybody on Oprah's show in 2000...... and Kerbango, for anybody who wanted an internet radio, that had the sound quality of a clock radio) The entire 3Com campus was HUGE, I believe it's Marvell now. Good times, good times

[Paul Hovnanian]

Won't work?

Back in those days, I can remember the 'Ethernet won't work' pitches. I also remember the TCP won't work, HTTP/HTML won't work, Windows-OS/2-Unix-whatever won't work. Blah, blah, blah.

The most interesting thing was that it was always the same group of people at my company that ran around with these pitches. Some sort of anti-marketing shills that vendors knew would take a few bucks to spread their propaganda white papers around. Pretty soon, the smarter folks started recognizing them and, whenever they came around, there were always coughs of "Bullsh*t!" behind our hands.

Some of them are still around and at least one major prominent company appears to be dealing with the fallout of bad engineering decisions pushed down from the MBA crowd. Some people never learn.

[Pet Peeve]

Re: Won't work?

Oh, heck yes. The network guys that wired our new building with token ring did it totally because it could run SNA, and would save a fortune on cabling the mainframe printers that were all over the building.

SNA was so overloaded with adminium that most people just said "forget it" when there was a request for a new terminal or printer in an area - they just made do walking a little farther or hanging out in the "tube room" all day instead of their desks. Even once we had the ring net set up, it was a headache to get changes made. Another thing I remember was when PCs first started getting used to replace 3278/9s, you had to buy a special network card to hook up to the old coax net, because the SNA gods refused to put workstations on the token ring SNA bridge. When they yanked all that crap out a few years later and put in ethernet, it was a very good day.

[Big_Friendly]

10 > 4

I saw RM speak in Toronto about 1989. He wrote 10 >4 on the whiteboard and began explained why Ethernet was better than Token Ring. At that time the metality was "You can't go wrong if you buy IBM". I remember one large internations spice company that went down the whole IBM route. AS 400, Token ring, OS/2 applications and tons of custom development. They lost millions and finally ripped it all out except for the AS400 on Ethernet and Windows on PCs.

[PghMike]

IBM's standards setting abilities

Ethernet no doubt contributed to IBM's loss of standards setting abilities, but I always thought that the real hit was when IBM couldn't come out with a 386-based system fast enough, and Compaq took over defining what an IBM PC-compatible system was.

That being said, IBM's token ring, and associated SNA technology, certainly deserves the scorn heaped upon it. My first full-time job was writing a driver so that a Sun 2's serial chip could operate in SDLC mode so we could send LU6.2 verbs to an IBM printer, and thus print on an IBM SNA-based printer from a Sun 2 workstation.

Why? Because our University got free IBM printers at the time.

[Roland6]

Re: IBM's standards setting abilities

Yes it's odd looking back now how IBM believed that it could go 802.5'ish and the world would jump off the Ethernet v2 bandwagon (which was morphing into the 802.3 Ethernet bandwagon) and follow it's lead.

Yes 802.5 (like 802.4) has some good features, but for most deployments the benefits didn't justify the costs.

I'm not so sure about the specific lack of a 386-based (PC) system, as the total lack of any real desktop solution to compete with Sun, Digital, Apollo, Torch, SGI, etc. etc. all of which shipped with Ethernet (and TCP/IP) as standard. But I agree IBM very quickly lost the ability to define what "IBM PC-compatible" meant - any one remember MCA?

[Tim99]

You try and tell the young people of today that..

...and they won't believe you (4 Yorkshire men, Monty Python).

I lost skin, literally, on Token Ring, Ethernet (the major contenders) and ARCnet/Star net.

Back in the day, just as the IBM PS2 was shipping, we had a couple of IBM PC-LAN/ LAN Manager networks that used Token Ring. The performance was generally very good on things like shared database access. The bad points were the cost of the cards, cabling and hubs; and the thickness of the cabling (It looked as though you moor a boat with it). We had a "proper" installation done by our telecoms/networking engineers over a weekend to replace the ad-hoc taped cables snaking around the building that we had put in to get the system working. On Monday, nothing worked - The engineers had used normal twisted pair POTS cabling. When asked why, the lead engineer said "Oh, your thick IBM stuff is very difficult to install, it wont even bend around corners. It's just twisted pair - Telephone cable is twisted pair, so it doesn't get interference." - A week later it was replaced with proper cabling, but they were the last Token Ring systems we could get our internal engineers to install.

10BASE2 - Really easy to set up. No hub needed, you could use 10BASE5 trunk to get longer lengths than 600ft. The bad news was that the cheap Netware NE1000/2000 cards that we used on client PCs tended to jabber and flood the network with malformed packets - So, Bob Metcalf, collisions WERE a problem. Other "features" of 10BASE2 networks were office staff moving a desk and taking down the entire network by damaging or unplugging the cable; terminating one end by connecting the cable directly to the card without using a T-piece; and noise from not grounding a terminator (or grounding both terminators and getting an earth loop).

A simple cost/performance benefit would generally come out in favour of Ethernet. In the end we had standardized on DEC, *NIX, Netware and PC equipment, so other than having to connect to the odd large IBM system, cheap Ethernet worked fine.

Shouty icon... Old Ethernet engineers know why.

[Ian 55]

Re: You try and tell the young people of today that..

.. and they won't believe it was At Last the 1948 Show it featured on...

[Anonymous Coward]

IMHO, Ethernet *really* got usable with switching.

Naturally, some fairly shouty protocols sometimes undid the traffic segmentation (hello, Microsoft) but in general switching has propelled Ethernet along as a fairly happy medium.

The only challenge is segment surveillance (for network intrusion detection etc), you need a SPAN port to see it all..

[Christian Berger]

Re: IMHO, Ethernet *really* got usable with switching.

Well I wouldn't say "usable", but switching certainly turned Ethernet from "it's cheap" to "it's cheap and it works quite well".

I still remember my first workplace. They had some fairly nifty managed hubs, which could be segmented and connected to the switch we had. If only someone would have taken the time to properly re-wire the patch cables, there would have been a 8 fold increase in performance. But still, 20 computers hammering away on 10 MBit shared with file shares and dBase databases still worked acceptably well.

[roselan]

I love this kind artcile

thank you.

And I can't help to draw parallels with today situation.

In my youth. I thought token ring was better, because it was more organized and fair (to the clients). And that coax thing. Oh boy!

[Jock in a Frock]

We're a bunch of bandwidth-hungry buggers now!

I've recently installed some 10GbE LAN-PHY rings for a client so their designers can use realtime CAD modelling tools between three sites in the Midlands. They are using this 10GbE directly to their workstations.

Hopefully they are hosting torrent files too......

[Christian Berger]

Well the CCC is hosting an anual congress

And they usually have around 20-30 Gigabits per second Internet connectivity. Despite of having around 6k people now (each one coming with at least a notebook) they never manage to fill their line.

[Anonymous Coward]

IBM RS/6000 Marketing Meet

We'd started selling the new RS/6000s and I went to a tech briefing. The guy half giggled and half blushed as he explained, yes, this IBM box came with Ethernet as standard.

[Daniel B.]

Re: IBM RS/6000 Marketing Meet

Though it isn't the Ethernet we all think about these days. It had a fugly AUI port instead of either coax or RJ45. I managed our sole remaining RS/6000 workstation at our college computer lab during the early 2000's and the AUI-to-cat5 transceiver was constantly being fought for between the RS/6000 users and the CCNA dudes (the Cisco routers only had AUI ports). Yeech!

[Anonymous Coward]

Re: IBM RS/6000 Marketing Meet

"and the AUI-to-cat5 transceiver was constantly being fought for"

In recent years I kept a stock of discarded AUI 10base-T "fag packets" in my drawer. Every so often someone would move an essential old machine off its old AUI cable - and find they couldn't connect it to the new network. I even kept some short AUI cables - as sometimes even a small "fag packet" proved too bulky to attach directly.

[Matt Bradley]

Oh dear

I was already starting to dislike this guy by page 2 of the interview. Then I got to the bottom of page 3...

[Anonymous Coward]

Token Ring 4/16

I started out in my networking career mid 90s at Exxon Chemical. TR was the game with SNA and IBM mainframes.

I have spent the last 10 years re-designing ISP networks away from SDH to Ethernet-only (not Ethernet over SDH) and MPLS P2P or VPLS.

Good work!

[TheWeenie]

My first ever job was for an insurance company. OS/2 running SNA over Token Ring. I still remember the feeling of panic stood in front of a rack full of Madge ringswitches, listening to the "Bzzzzzzt" of the relays as the whole lot came crashing down! Back at the time we were ordering in Compaq hardware with built-in 10/100 PHY and then spending another £150 on a Token Ring card for each device. Back then a 10/100 Ethernet switch (unmanaged, naturally) would cost about the same as one card. We tried to set up a PoC in the office but couldn't source any money for a router.

Anyone remember HSTR? 100Mbit - cutting edge back then. Just a shame it used to fail about once a month.

[Anonymous Coward]

"History is written by the winners"

Token ring had the advantage that it was a ring so a single cable break was not a problem, E'net was, (way back), less resilient, backbone needed thick E'net cables (near 1cm diameter) with a mimimum curve radius of - don't remember, 1 metre? Signal distance was a big constraint too.

When E'net went twisted pair and jumped to 100Mbit T'ring was too late with bringing on competitive speed.

I heard a great explanation of T'ring vs E'net: In character (not in actual origin) T'ring was British, E'net American. The "British" T'ring node would have a conversation like: The token has arrived, I am invited to speak, but I have nothing to say so I'll pass the opportunity (and the decanter of Port) to my neighbour at the dinner table. The "American" E'net nodes all just shout as soon as an idea comes into their heads and the one that shouts loudest wins.

[KA]

Re: "History is written by the winners"

T-R was logically a ring, but wired as hub-and-spoke. When a spoke was broken the hub had to detect the problem and bypass that node in order to maintain the logical ring integrity.

[sabba]

Ah token-ring...

...the One True Ring (My Precious)!!

[TeeCee]

Killing token-ring.

All well and good, but overlooks one very important point. Ethernet never killed token-ring, IBM did that for them.

The chipset problem was alluded to in the article. The problem here was that if you wanted to use any of the IBM software (e.g. the PC terminal emulators) you had to use cards with the honest-to-god, gen-u-ine IBM TROPIC chipset (that's the big, square metal thing on an IBM card). Nothing else would do. For all other purposes, third-party cards would do just fine.

IBM, in their infinite wisdom, only ever produced "real" IBM cards as 8 bit ISA cards which, from the price, seemed to be assembled from Unicorn vellum laid on an Unobtanium substrate. Eventually they decided to license TROPIC and the result was the excellent (and sensibly priced) 3com Tokenlink III. As usual with IBM, it was too little too late and Ethernet had its feet firmly under the table by then.

I'm afraid that all the alleged FUD was actually, er, true. You only needed four PCs running DOOM and chaingunning the fuck out of each other to prove that 4meg token-ring shat all over Ethernet for throughput under load. TR gave you a usable (if slow) network with that going on, Ethernet cacked itself on the spot. One of TRs best features was that it degraded gracefully under heavy load. 16meg (that's 16, not 12!!) TR just flew in comparison to Ethernet.

Token-ring had diagnostics and redundancy built in, which is why the cards were pricier than their Ethernet counterparts. Yes, it was entirely possible to have the thing give up without your having a clue what had gone wrong, but if you weren't running something with the capability to diagnose and act on beacon frames and you hadn't made that the designated ring controller, it was entirely your fault! The concept of the token "falling out" is actually an old joke. If you do manage to axe the machine that has the token at the time, it's NAUN (Nearest Active Upstream Neighbour) spots the problem and generates a new token containing a beacon frame to advise the ring controller of the change. This happens in milliseconds, not minutes.

There certainly is a load of FUD around all this, but he's the one pushing it!

[Peter Simpson 1]

Re: Killing token-ring.

IMHO, the thing that killed Token Ring was the cost of the license for the MAC code and the fact that it couldn't run on CAT5. Whereas Ethernet was truly asynchronous and had standardised signalling waveforms carefully designed to work on unshielded TP, Token Ring never quite got to that point (as the clock jitter requirement precluded bandwith limited waveforms on the wire).

I spent much time at 3Com, trying to get 16 meg Token Ring switches to work reliably at maximum cable lengths (and then to get them to pass FCC emissions testing). IMHO, it was never to be. Meanwhile, Ethernet over twisted pair was heading for 100 megabit/sec. Game over.

//I did manage to implement a parser for the source routing field in a CPLD, so there was that, at least...

[TeeCee]

Re: Killing token-ring.

No "cost per MAC", see answer above.

I ran 16meg TR over CAT5[1]. The cable length specs were conservative, to say the least. This only ever failed once. Some contractors moved an office wall and spliced(!) the CAT5 that ran along it. My reactions on finding the splice were unprintable. Suffice to say they revolved around the contractors, their parentage, their uncanny resemblance to various parts of human anatomy and what I'd do to them if I ever got them in a room with a fire axe.

[1] 3com cards and Andrews' MAUs IIRC.

[Ken 16]

Bring back the Vampire Tap!

That was the only fun part of networking.

[Charles-A Rovira]

I used to work for an outfit that used token ring.

They ceased to exist soon after that.

The joke around the office was sending each newbies looking for the "dropped packet" in the token ring.

Yeah, they were that lame...