I found the reddit thread referenced in the video and left comments -
On the original ethernet being a mistake -
"I have another perspective on this - the bus concept at its roots dates to 1974/75 - a machine that could function as a hub in 1975 or even in 1979 would have potentially been as large and costly as the mini-computers Ethernet was intended to interconnect. A switch at that time would have been extraordinary costly.
Bus was the right technology for the time (as in the right one to go to market), it was right because it was an easy low cost means to get the LAN concept into more peoples hands so they could see the utility of it.
What I did find surprising was Metcalfe's resistance to UTP based standards with a hub in the middle once the technology did catch up - my gut from here 40 years on - makes me think that he was concerned about protecting his investments in 10base5 and 10base2, and making sure his installed base didnt feel like they had an obsolete product - as well as protecting 3Com's position as a 'Market Leader' in the space.
As an aside, I remember working on 10Base2 networks well into the late 90's and early 2000's - it was widely used in certain situations - like computer labs - where the bus topology made installation significantly easier than home runs would have been."
On 8P8C being picked as the connector -
"Also as someone who has spent a portion of my career doing cabling, thank you for picking the 8P8C modular system, its significantly better - both in cost and ease of assembly than anything I'm aware of existing at that time.
While there are some electrical issues with the 8P8C, the connector has been flexible enough that with minor design changes to keep up as line speeds have increased.
As a note, there are several vendors that make push-thru modular connectors (as in the wire extends beyond the front of the connector shell and the crimper trims it, much like how the cut blade on a punch tool works), which completely eliminate the need to accurately trim and face the wire ends - a huge time/frustration savings."
486sx33 2 days ago [-]
I worked on 10Base2 as well in the mid to late 90s and it was a pain. The terminators, the length restrictions, the repeaters… no hubs but it was hell, one section damaged the whole thing goes down. 10mbit limit to the entire combined throughput. Need more? Add another card to the server… want to talk from one loop to another? Watch server cpu
Kelvin506 2 days ago [-]
I made a good living in the 90s going to orgs with "network problems" and fixing things. One of the most common problems was a coax ethernet network that had outgrown itself.
The fix was installing a 10bT/100bTX switch, installing 100bTX NICs in the servers, and breaking up their 10b2/10b5 networks into smaller segments each with media converter connecting it to the switch. An unmanaged 16/24-port 10/100 switch was "only" a couple hundred bucks back then.
Aloha 2 days ago [-]
It did work really well for a ton of computers in the same room in a line however.
leoc 2 days ago [-]
Also (as you know), twisted-pair Ethernet with a hub (let alone that STARNET daisy-chain thingy!) just isn't that different from coax Ethernet. Electrically it's largely shared-medium, half-duplex bus Ethernet wearing a mechanical hub-and-spoke disguise, and the hub is by definition just a multiport repeater. It's only when the hub is replaced by a multiport switch that shared-medium Ethernet is decisively left behind.
Which brings me to one thing that puzzled me about the video. In the part starting from 10:37 https://youtu.be/f8PP5IHsL8Y?t=637 it describes STARLAN as specifying two pairs of wires, one for transmit and one for receive. But naturally you can't really have full duplex without a switch at the other end of the wire; and yet the STARLINK "Network Extension Unit" is described as a hub (and indeed it seems that the first Ethernet switches/multiport bridges didn't ship until 1989-1990). So when at 12:10 https://youtu.be/f8PP5IHsL8Y?t=730 the video seems to suggest that STARLINK being half-duplex was a step back from the original plan, I don't see how that can be the case: I assume that first-generation STARLINK was never intended to provide full duplex, and the two pairs of wires were just reserved for future upgrades. That seems compatible with what Richard Bennett says from 12:29 . That's still interesting, though, because it suggests that the STARLINK contributors must have had switches in mind back in 1983.
Aloha 2 days ago [-]
I very much agree - it has the advantage of not being an electrically shared medium however, which is one of the things that made it superior to 10Base2 or 10Base5
Here is some interesting documentation from the original StarLAN Hub -
Yep. Very fond of the "EZ RJ 45" line in particular. The strain relief that's built into them does not interfere with loading the cable and actually gets a good bite into the jacket once crimped.
The other advantage I've found is that it's harder to accidentally make defective cable ends. Trying to trim and face yourself you run the risk of one of the lines being too short or pulled back during the crimp and not mating well with the lug. The EZ RJ 45 effectively eliminates this as there is always wire under the lug during the crimp.
Aloha 9 hours ago [-]
Trim and face is the part thats really hard to learn, and do consistently well - also if you do trim and face well, you're likely untwisting too much wire to get it aligned correctly so it doesn't roll over in the connector while putting it on.
EZ RJ 45 does solve all of these issue, and makes for a reliable cable 100% of the time.
zabzonk 2 days ago [-]
My history with networking:
- late 70s, rs232 terminals. well, it kind of worked. i spent a lot of time under the desk
- early 80s, frozen hose ethernet - i remember not being able to get a sun workstation to sit on the desk because the cable lifted it off too much
- late 80s coax ethernet all sorts of termination and other problems
- 90s twisted pair and hubs - things started working as they should.
- a bit later cheap switches came along, but still lots of wires at the desk
- now, wireless bliss.
- don't talk to me about token ring.
But of course I don't do this stuff any more.
pcdoodle 2 days ago [-]
Never heard of frozen hose, funny.
I briefly used "thin" coax before the transition to IP cameras, it was a basic NTSC signal into a PCI capture card. 12/24V was pumped over a secondary cable attached to the main coax (siamese)
We were digital plumbers!
elisiariocouto 2 days ago [-]
The Serial Port is a very good channel. Been learning a lot with their last videos, almost feels like we're traveling in time :)
loup-vaillant 2 days ago [-]
Dammit, the video doesn’t explain why we have 4 twisted pairs in our cables instead of just 2! One for Rx, one for Tx, but then why did they keep the other two? Surely they didn’t have 4 dead wires just because of legacy? So why the additional pairs, what were they used for?
dfox 1 days ago [-]
Originally it was more or less one for POTS, one Rx, one Tx and one spare (used for power in ISDN S/T interface, which otherwise shares the same pinout).
Then came 1000-base-T, where the “base” is kind of misnomer and it is actually related to SHDSL. Each pair is essentially a separate full-duplex 250Mbps link with active echo cancellation and four of them is combined to create one 1Gbps link.
tuetuopay 2 days ago [-]
Signal integrity basically. Using more pairs means reducing the bandwidth needed on each pair, thus the cost of the cable.
userbinator 2 days ago [-]
It's also a tradeoff with the cost of the NIC hardware --- look at how much bandwidth xDSL technologies have been able to squeeze into a single traditional twisted-pair phone line, but they use far more complex signal processing techniques than Ethernet.
1 days ago [-]
2 days ago [-]
karanveer 6 days ago [-]
just saw this recommendation on yt 1 minute ago.. released 3 hours ago by The Serial Port...small world.
kristofferR 2 days ago [-]
That doesn't make any sense, this video was published days ago. I watched it then.
jasode 2 days ago [-]
Unfortunately, when submissions get put back into a "2nd chance queue", the "elapsed time calc" is modified to be more recent -- which causes confusion. The thread that's given "new life" for more engagement now has a mix of real and fake "x hours ago".
If you're on a desktop browser instead of a smartphone, you can use F12 Dev Tools to inspect the parent post[1] elements to see that the original timestamp is still there as "2024-12-26T05:23:25 1735190605" ... which is actually about 4 days ago.
Some search engines (bing) still has a cache of this thread we're in from 4 days ago.
The comment you replied to was published 4 days ago
Thorrez 2 days ago [-]
It says "9 hours ago" to me. Is this one of those situations where HN reposted the post and the comments as well? I wish the date and time were kept accurate instead of being confusingly changed.
On the original ethernet being a mistake -
"I have another perspective on this - the bus concept at its roots dates to 1974/75 - a machine that could function as a hub in 1975 or even in 1979 would have potentially been as large and costly as the mini-computers Ethernet was intended to interconnect. A switch at that time would have been extraordinary costly.
Bus was the right technology for the time (as in the right one to go to market), it was right because it was an easy low cost means to get the LAN concept into more peoples hands so they could see the utility of it.
What I did find surprising was Metcalfe's resistance to UTP based standards with a hub in the middle once the technology did catch up - my gut from here 40 years on - makes me think that he was concerned about protecting his investments in 10base5 and 10base2, and making sure his installed base didnt feel like they had an obsolete product - as well as protecting 3Com's position as a 'Market Leader' in the space.
As an aside, I remember working on 10Base2 networks well into the late 90's and early 2000's - it was widely used in certain situations - like computer labs - where the bus topology made installation significantly easier than home runs would have been."
On 8P8C being picked as the connector -
"Also as someone who has spent a portion of my career doing cabling, thank you for picking the 8P8C modular system, its significantly better - both in cost and ease of assembly than anything I'm aware of existing at that time.
While there are some electrical issues with the 8P8C, the connector has been flexible enough that with minor design changes to keep up as line speeds have increased.
As a note, there are several vendors that make push-thru modular connectors (as in the wire extends beyond the front of the connector shell and the crimper trims it, much like how the cut blade on a punch tool works), which completely eliminate the need to accurately trim and face the wire ends - a huge time/frustration savings."
The fix was installing a 10bT/100bTX switch, installing 100bTX NICs in the servers, and breaking up their 10b2/10b5 networks into smaller segments each with media converter connecting it to the switch. An unmanaged 16/24-port 10/100 switch was "only" a couple hundred bucks back then.
Which brings me to one thing that puzzled me about the video. In the part starting from 10:37 https://youtu.be/f8PP5IHsL8Y?t=637 it describes STARLAN as specifying two pairs of wires, one for transmit and one for receive. But naturally you can't really have full duplex without a switch at the other end of the wire; and yet the STARLINK "Network Extension Unit" is described as a hub (and indeed it seems that the first Ethernet switches/multiport bridges didn't ship until 1989-1990). So when at 12:10 https://youtu.be/f8PP5IHsL8Y?t=730 the video seems to suggest that STARLINK being half-duplex was a step back from the original plan, I don't see how that can be the case: I assume that first-generation STARLINK was never intended to provide full duplex, and the two pairs of wires were just reserved for future upgrades. That seems compatible with what Richard Bennett says from 12:29 . That's still interesting, though, because it suggests that the STARLINK contributors must have had switches in mind back in 1983.
Here is some interesting documentation from the original StarLAN Hub -
https://vtda.org/docs/computing/AT&T/AT&T_Starlan_10_Hub.pdf
There is also some neat and ancient documentation still lovingly preserved on the NCR website -
https://onlinehelp.ncr.com/Retail/Workstations/Wiring/Ethern...
Yep. Very fond of the "EZ RJ 45" line in particular. The strain relief that's built into them does not interfere with loading the cable and actually gets a good bite into the jacket once crimped.
The other advantage I've found is that it's harder to accidentally make defective cable ends. Trying to trim and face yourself you run the risk of one of the lines being too short or pulled back during the crimp and not mating well with the lug. The EZ RJ 45 effectively eliminates this as there is always wire under the lug during the crimp.
EZ RJ 45 does solve all of these issue, and makes for a reliable cable 100% of the time.
- late 70s, rs232 terminals. well, it kind of worked. i spent a lot of time under the desk
- early 80s, frozen hose ethernet - i remember not being able to get a sun workstation to sit on the desk because the cable lifted it off too much
- late 80s coax ethernet all sorts of termination and other problems
- 90s twisted pair and hubs - things started working as they should.
- a bit later cheap switches came along, but still lots of wires at the desk
- now, wireless bliss.
- don't talk to me about token ring.
But of course I don't do this stuff any more.
I briefly used "thin" coax before the transition to IP cameras, it was a basic NTSC signal into a PCI capture card. 12/24V was pumped over a secondary cable attached to the main coax (siamese)
We were digital plumbers!
Then came 1000-base-T, where the “base” is kind of misnomer and it is actually related to SHDSL. Each pair is essentially a separate full-duplex 250Mbps link with active echo cancellation and four of them is combined to create one 1Gbps link.
If you're on a desktop browser instead of a smartphone, you can use F12 Dev Tools to inspect the parent post[1] elements to see that the original timestamp is still there as "2024-12-26T05:23:25 1735190605" ... which is actually about 4 days ago.
Some search engines (bing) still has a cache of this thread we're in from 4 days ago.
https://imgur.com/a/example-hacker-news-thread-put-back-into...
If one were to click on that 4 day old thread, you come here where all the comments are "x hours ago". It's confusing.
[1] https://news.ycombinator.com/item?id=42546687