I have lots of "1-to-1" connections between network appliances.
I'm trying to think of the most effective (not efficient!) way
of doing so.

E.g., imagine two switch-like appliances (lots of ports arranged
side by side). They have to be connected to each other.

One approach is to locate them physically adjacent and use identical
length cords to connect port 1 to 1, 2 to 2, 3 to 3, etc. This works
as long as 1 ALWAYS connects to 1 and never, e.g., 18!

When you have more than two such devices, there are other options.
E.g., imagine a set of 8.

You can interleave them: AaBbCcDd and try tot same 1-to-1 connection.
With the same caveat.

Or, could create "sandwiches" where half of A goes to half of its
counterpart "above" with the other half to the counterpart "below".
Esp for devices with two or more rows of 8P8Cs.

You could lump all of the devices of one type together and
those of their counterparts: ABCDabcd

Or, not exploit proximity and just put them where they fit: AabcBCDd

Its an issue because you have a lot of potential "tangle" as well as
gaining physical access to ports that may end up buried in a ratsnest.
The datacenter solution is just to dress the cables off to the side
and keep the connection faces relatively exposed. That comes at the
expense of more cable and the difficulty of extracting/inserting a
cable from that neat bundle. (Recall that I have to address users
who may be blind or physically disabled.)

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

Don Y <blockedofcourse@foo.invalid> wrote:

I have lots of "1-to-1" connections between network appliances.
I'm trying to think of the most effective (not efficient!) way
of doing so.

E.g., imagine two switch-like appliances (lots of ports arranged
side by side). They have to be connected to each other.

One approach is to locate them physically adjacent and use identical
length cords to connect port 1 to 1, 2 to 2, 3 to 3, etc. This works
as long as 1 ALWAYS connects to 1 and never, e.g., 18!

When you have more than two such devices, there are other options.
E.g., imagine a set of 8.

You can interleave them: AaBbCcDd and try tot same 1-to-1 connection.
With the same caveat.

Or, could create "sandwiches" where half of A goes to half of its
counterpart "above" with the other half to the counterpart "below".
Esp for devices with two or more rows of 8P8Cs.

You could lump all of the devices of one type together and
those of their counterparts: ABCDabcd

Or, not exploit proximity and just put them where they fit: AabcBCDd

Its an issue because you have a lot of potential "tangle" as well as
gaining physical access to ports that may end up buried in a ratsnest.
The datacenter solution is just to dress the cables off to the side
and keep the connection faces relatively exposed. That comes at the
expense of more cable and the difficulty of extracting/inserting a
cable from that neat bundle. (Recall that I have to address users
who may be blind or physically disabled.)

The 'sandwich' arrangement will be easiest for a blind person because
they only have to remember one linear order of the ports:

ABCDEFG...
abcdefg...

There is a risk with the 'sandwich' system that someone could plug an
output to another output in the same row.

Another possibility would be to use a matrix with shorting plugs and no connecting leads. If it is imperative that only one-to-one connections
are permitted, the sockets could be break jacks to interrupt the
connection to all the subsequent jacks in that row or column. (The disadvantage of break jacks is that there are a lot of series contacts
to go faulty.)


--
~ Liz Tuddenham ~
(Remove the ".invalid"s and add ".co.uk" to reply)
www.poppyrecords.co.uk

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On 7/2/2025 2:19 AM, Liz Tuddenham wrote:

I have lots of "1-to-1" connections between network appliances.
I'm trying to think of the most effective (not efficient!) way
of doing so.

E.g., imagine two switch-like appliances (lots of ports arranged
side by side). They have to be connected to each other.

One approach is to locate them physically adjacent and use identical
length cords to connect port 1 to 1, 2 to 2, 3 to 3, etc. This works
as long as 1 ALWAYS connects to 1 and never, e.g., 18!

When you have more than two such devices, there are other options.
E.g., imagine a set of 8.

You can interleave them: AaBbCcDd and try tot same 1-to-1 connection.
With the same caveat.

Or, could create "sandwiches" where half of A goes to half of its
counterpart "above" with the other half to the counterpart "below".
Esp for devices with two or more rows of 8P8Cs.

You could lump all of the devices of one type together and
those of their counterparts: ABCDabcd

<https://mega.nz/file/omoiWZoD#IXzd5heL6QmTIZJ5upl-RpuPDQzXODugVb9IGJzZ0RI>

The 'sandwich' arrangement will be easiest for a blind person because
they only have to remember one linear order of the ports:

ABCDEFG...
abcdefg...

I think so. Cables never cross so you don't have to trace a cable under/behind another (like in the interleaved approach).

The downside is the top row of the top device and bottom row of the bottom device are missing adjacent mates.

A possible solution is to build one of the devices as a "single row"
(instead of the dual row common in switches). This would require placing
one above/below each row of an opposite device. And, thus ensures the
top and bottom devices have nearby mates.

[This seems to be a good take-away!]

There is a risk with the 'sandwich' system that someone could plug an
output to another output in the same row.

I think they can be relied upon to "follow" the adjacent patch cords
to know when they have advanced or fallen to another row.

Another possibility would be to use a matrix with shorting plugs and no connecting leads. If it is imperative that only one-to-one connections
are permitted, the sockets could be break jacks to interrupt the
connection to all the subsequent jacks in that row or column. (The disadvantage of break jacks is that there are a lot of series contacts
to go faulty.)

I want to allow for the possibility -- in an unusual situation -- for
1 to be broken from 1 and mated to 82, instead. E.g., if the mated 1 malfunctions.

But, this would be a "singleton" connection that could easily stand out
among the more orderly other connections.

The sandwich illustrated relies on lots of very short patch cords.
This keeps things dressed nice and tidy. But, may require care in
their manufacture -- e.g., if you plug in one end and have only
a few inches of service loop to ensure the other end is *oriented*
in the correct direction.

It also means you'd need a few "exception handlers" -- longer cords
to deal with the 1-to-82 case alluded to above.

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

Don Y <blockedofcourse@foo.invalid> wrote:

On 7/2/2025 2:19 AM, Liz Tuddenham wrote:

I have lots of "1-to-1" connections between network appliances.
I'm trying to think of the most effective (not efficient!) way
of doing so.

E.g., imagine two switch-like appliances (lots of ports arranged
side by side). They have to be connected to each other.

One approach is to locate them physically adjacent and use identical
length cords to connect port 1 to 1, 2 to 2, 3 to 3, etc. This works
as long as 1 ALWAYS connects to 1 and never, e.g., 18!

When you have more than two such devices, there are other options.
E.g., imagine a set of 8.

You can interleave them: AaBbCcDd and try tot same 1-to-1 connection.
With the same caveat.

Or, could create "sandwiches" where half of A goes to half of its
counterpart "above" with the other half to the counterpart "below".
Esp for devices with two or more rows of 8P8Cs.

You could lump all of the devices of one type together and
those of their counterparts: ABCDabcd

<https://mega.nz/file/omoiWZoD#IXzd5heL6QmTIZJ5upl-RpuPDQzXODugVb9IGJzZ0RI

The 'sandwich' arrangement will be easiest for a blind person because
they only have to remember one linear order of the ports:

ABCDEFG...
abcdefg...

I think so. Cables never cross so you don't have to trace a cable under/behind another (like in the interleaved approach).

The downside is the top row of the top device and bottom row of the bottom device are missing adjacent mates.

A possible solution is to build one of the devices as a "single row"
(instead of the dual row common in switches). This would require placing
one above/below each row of an opposite device. And, thus ensures the
top and bottom devices have nearby mates.

[This seems to be a good take-away!]

There is a risk with the 'sandwich' system that someone could plug an output to another output in the same row.

I think they can be relied upon to "follow" the adjacent patch cords
to know when they have advanced or fallen to another row.

Another possibility would be to use a matrix with shorting plugs and no connecting leads. If it is imperative that only one-to-one connections
are permitted, the sockets could be break jacks to interrupt the
connection to all the subsequent jacks in that row or column. (The disadvantage of break jacks is that there are a lot of series contacts
to go faulty.)

I want to allow for the possibility -- in an unusual situation -- for
1 to be broken from 1 and mated to 82, instead. E.g., if the mated 1 malfunctions.

But, this would be a "singleton" connection that could easily stand out
among the more orderly other connections.

The sandwich illustrated relies on lots of very short patch cords.
This keeps things dressed nice and tidy. But, may require care in
their manufacture -- e.g., if you plug in one end and have only
a few inches of service loop to ensure the other end is *oriented*
in the correct direction.

It also means you'd need a few "exception handlers" -- longer cords
to deal with the 1-to-82 case alluded to above.

It sounds as though your problem is very similar to the problems faced
by the designers of manual telephone switchboards a century ago. I
would look around and see if it was possible to pick up some NOS
switchboards - or at least read up on some of the old journals: Bell
Labs, POEEJ.


--
~ Liz Tuddenham ~
(Remove the ".invalid"s and add ".co.uk" to reply)
www.poppyrecords.co.uk

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On 7/2/2025 12:01 PM, Liz Tuddenham wrote:

Don Y <blockedofcourse@foo.invalid> wrote:

The 'sandwich' arrangement will be easiest for a blind person because
they only have to remember one linear order of the ports:

ABCDEFG...
abcdefg...

I think so. Cables never cross so you don't have to trace a cable
under/behind another (like in the interleaved approach).

The downside is the top row of the top device and bottom row of the bottom >> device are missing adjacent mates.

A possible solution is to build one of the devices as a "single row"
(instead of the dual row common in switches). This would require placing
one above/below each row of an opposite device. And, thus ensures the
top and bottom devices have nearby mates.

[This seems to be a good take-away!]

There is a risk with the 'sandwich' system that someone could plug an
output to another output in the same row.

I think they can be relied upon to "follow" the adjacent patch cords
to know when they have advanced or fallen to another row.

Another possibility would be to use a matrix with shorting plugs and no
connecting leads. If it is imperative that only one-to-one connections
are permitted, the sockets could be break jacks to interrupt the
connection to all the subsequent jacks in that row or column. (The
disadvantage of break jacks is that there are a lot of series contacts
to go faulty.)

I want to allow for the possibility -- in an unusual situation -- for
1 to be broken from 1 and mated to 82, instead. E.g., if the mated 1
malfunctions.

But, this would be a "singleton" connection that could easily stand out
among the more orderly other connections.

The sandwich illustrated relies on lots of very short patch cords.
This keeps things dressed nice and tidy. But, may require care in
their manufacture -- e.g., if you plug in one end and have only
a few inches of service loop to ensure the other end is *oriented*
in the correct direction.

It also means you'd need a few "exception handlers" -- longer cords
to deal with the 1-to-82 case alluded to above.

It sounds as though your problem is very similar to the problems faced
by the designers of manual telephone switchboards a century ago. I
would look around and see if it was possible to pick up some NOS
switchboards - or at least read up on some of the old journals: Bell
Labs, POEEJ.

I think the biggest takeaway is to design things that are "one row"
of 8P8C's. This gives you the most flexibility in interconnecting
them without "tangle" or "overlap".

Also, ensure there are no markings that dictate a particular
orientation. So, a column is just as feasible as a row.

I've already made provisions to facilitate *locating* specific
cables as well as *indicating* ones of interest.

Next, a reexamination as to whether 8P8Cs are the "right" connectors
(just because data centers use them doesn't make them suitable for
use in other applications!). The telephone "plugboard" would be
nice (no orientation problems) but may be an issue electrically.

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

Don Y <blockedofcourse@foo.invalid> wrote:

On 7/2/2025 12:01 PM, Liz Tuddenham wrote:

Don Y <blockedofcourse@foo.invalid> wrote:

The 'sandwich' arrangement will be easiest for a blind person because
they only have to remember one linear order of the ports:

ABCDEFG...
abcdefg...

I think so. Cables never cross so you don't have to trace a cable
under/behind another (like in the interleaved approach).

The downside is the top row of the top device and bottom row of the bottom >> device are missing adjacent mates.

A possible solution is to build one of the devices as a "single row"
(instead of the dual row common in switches). This would require placing >> one above/below each row of an opposite device. And, thus ensures the
top and bottom devices have nearby mates.

[This seems to be a good take-away!]

There is a risk with the 'sandwich' system that someone could plug an
output to another output in the same row.

I think they can be relied upon to "follow" the adjacent patch cords
to know when they have advanced or fallen to another row.

Another possibility would be to use a matrix with shorting plugs and no >>> connecting leads. If it is imperative that only one-to-one connections >>> are permitted, the sockets could be break jacks to interrupt the
connection to all the subsequent jacks in that row or column. (The
disadvantage of break jacks is that there are a lot of series contacts >>> to go faulty.)

I want to allow for the possibility -- in an unusual situation -- for
1 to be broken from 1 and mated to 82, instead. E.g., if the mated 1
malfunctions.

But, this would be a "singleton" connection that could easily stand out
among the more orderly other connections.

The sandwich illustrated relies on lots of very short patch cords.
This keeps things dressed nice and tidy. But, may require care in
their manufacture -- e.g., if you plug in one end and have only
a few inches of service loop to ensure the other end is *oriented*
in the correct direction.

It also means you'd need a few "exception handlers" -- longer cords
to deal with the 1-to-82 case alluded to above.

It sounds as though your problem is very similar to the problems faced
by the designers of manual telephone switchboards a century ago. I
would look around and see if it was possible to pick up some NOS switchboards - or at least read up on some of the old journals: Bell
Labs, POEEJ.

I think the biggest takeaway is to design things that are "one row"
of 8P8C's. This gives you the most flexibility in interconnecting
them without "tangle" or "overlap".

Also, ensure there are no markings that dictate a particular
orientation. So, a column is just as feasible as a row.

I've already made provisions to facilitate *locating* specific
cables as well as *indicating* ones of interest.

Next, a reexamination as to whether 8P8Cs are the "right" connectors
(just because data centers use them doesn't make them suitable for
use in other applications!). The telephone "plugboard" would be
nice (no orientation problems) but may be an issue electrically.

It looks as though a vertical panel with two rows in the same numerical
order, with the 'outs' above their corresponding 'ins' and joined by
very short interconnecting links in the 'normal' configuration would be
the best arrangement. Below that could be a horizontal projecting shelf
with much longer leads tensioned with weights so that they could be
pulled up and used to make non-standard interconnections. When released
they would retract neatly out of the way.

That was the system used in telephone manual switchboards and it worked
very well, with the interconnections being easy to trace and no tangles
of loose leads.

If you could justify the design and special manufacturing costs, you
could go one stage further and make it so that an ''out' and 'in' pair
of sockets on one 'channel' were automatically connected to each other
in the absence of a plug. That would do away with the short
interconnectors altogether. This method was used by the BBC for their apparatus racks, so that only the non-standard interconnections needed
external leads.

If standard Post Office Gauge'B' plugs (316-type) and leads could carry
the signals without degradation, there have been millions of standard ready-made jack strips with break-jacks already manufactured which will
now be lying around as surplus stock. (Grab them before the government
tries to buy them all back when they realise manual telephone
switchboards are EMP-proof.)


--
~ Liz Tuddenham ~
(Remove the ".invalid"s and add ".co.uk" to reply)
www.poppyrecords.co.uk

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

Don Y <blockedofcourse@foo.invalid> wrote:

I want to allow for the possibility -- in an unusual situation -- for
1 to be broken from 1 and mated to 82, instead. E.g., if the mated 1 malfunctions.

But, this would be a "singleton" connection that could easily stand out
among the more orderly other connections.

The sandwich illustrated relies on lots of very short patch cords.
This keeps things dressed nice and tidy. But, may require care in
their manufacture -- e.g., if you plug in one end and have only
a few inches of service loop to ensure the other end is *oriented*
in the correct direction.

It also means you'd need a few "exception handlers" -- longer cords
to deal with the 1-to-82 case alluded to above.

If you provision your installers with a bag of (say) 100x yellow 0.2m cables and a bag of a much smaller number of red 0.5m cables (let's say 4, in this example), the short cables will get used for the vertical links and the long red cables will get used for the links out of sequence. That way the ones
that are unordered will stand out and be easy to identify.

I expect that even the dumbest installer is not going to fit links 1-4 with
the long red cables and then links 5-80 with the short yellow cables. And there will be a certain physical limit to them crosswiring (port 1 to port 2 maybe, but not 1 to 20) because the short cables won't reach.

Theo

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

Next, a reexamination as to whether 8P8Cs are the "right" connectors
(just because data centers use them doesn't make them suitable for
use in other applications!). The telephone "plugboard" would be
nice (no orientation problems) but may be an issue electrically.

It looks as though a vertical panel with two rows in the same numerical order, with the 'outs' above their corresponding 'ins' and joined by
very short interconnecting links in the 'normal' configuration would be
the best arrangement.

If by "rows" you mean "columns", I think that is right.

Arrange the connectors to be adjacent with indicators "outboard".
This would make it easy for sighted/unsighted folks to get a quick
assessment of the state of everything:

O [] [] O O [] [] O O [] [] O
O [] [] O O [] [] O O [] [] O
O [] [] O O [] [] O O [] [] O
O [] [] O O [] [] O O [] [] O
O [] [] O O [] [] O O [] [] O

where:
[] connection point
O indicator

Below that could be a horizontal projecting shelf
with much longer leads tensioned with weights so that they could be
pulled up and used to make non-standard interconnections. When released
they would retract neatly out of the way.

I'm assuming any connection of that sort would be infrequent and
temporary. So, a single (?) overly long patch cord could do the
trick.

As it (and it alone) would skew across the panel, it should be easy
to identify as an exception. And, follow to each end.

That was the system used in telephone manual switchboards and it worked
very well, with the interconnections being easy to trace and no tangles
of loose leads.

If you could justify the design and special manufacturing costs, you
could go one stage further and make it so that an ''out' and 'in' pair
of sockets on one 'channel' were automatically connected to each other
in the absence of a plug. That would do away with the short
interconnectors altogether.

That may be possible. I had planned on making "singletons" that were effectively edge-stackable as the total number (and configuration)
would vary from system to system. "Standardizing" on some arbitrary
packaging (dimensions) would likely lead to complications at installation.

E.g., here, I would arrange things in columns of 36. Shorter columns
(e.g., 24) would necessitate more columns (i.e., greater width)
while longer columns would make them span longer reaches so the points
at one end were less accessible than the other.

This method was used by the BBC for their
apparatus racks, so that only the non-standard interconnections needed external leads.

If standard Post Office Gauge'B' plugs (316-type) and leads could carry
the signals without degradation, there have been millions of standard ready-made jack strips with break-jacks already manufactured which will
now be lying around as surplus stock. (Grab them before the government
tries to buy them all back when they realise manual telephone
switchboards are EMP-proof.)

They're (currently) Gbe drops. So, there is some care needed in their
physical termination. I suspect they will be replaced with CAT6 in
the not too distant future as Gbe is becoming "old hat". That may
pose some problems if the connection points are located too closely together (minimum bend radius)

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On 7/3/2025 8:36 AM, Theo wrote:

Don Y <blockedofcourse@foo.invalid> wrote:

I want to allow for the possibility -- in an unusual situation -- for
1 to be broken from 1 and mated to 82, instead. E.g., if the mated 1
malfunctions.

But, this would be a "singleton" connection that could easily stand out
among the more orderly other connections.

The sandwich illustrated relies on lots of very short patch cords.
This keeps things dressed nice and tidy. But, may require care in
their manufacture -- e.g., if you plug in one end and have only
a few inches of service loop to ensure the other end is *oriented*
in the correct direction.

It also means you'd need a few "exception handlers" -- longer cords
to deal with the 1-to-82 case alluded to above.

If you provision your installers with a bag of (say) 100x yellow 0.2m cables and a bag of a much smaller number of red 0.5m cables (let's say 4, in this example), the short cables will get used for the vertical links and the long red cables will get used for the links out of sequence. That way the ones that are unordered will stand out and be easy to identify.

You don't worry about an "installer" -- he'd (supposedly) be cosen
for his competencies.

What you worry about is the *user* making changes to the wiring,
typically because he sees something undesirable happening,
wants to make a change or has been directed to make a *specific*
change (by the system or by a remote troubleshooter).

E.g., "move the A end of cable 137 to the unused A connection 205."
Or, "swap the B ends of cables 128 and 52."

He would have to be able to execute this correctly with no
technical training -- possibly blind, possibly confined to a wheelchair,
etc.

I expect that even the dumbest installer is not going to fit links 1-4 with the long red cables and then links 5-80 with the short yellow cables. And there will be a certain physical limit to them crosswiring (port 1 to port 2 maybe, but not 1 to 20) because the short cables won't reach.

But *users* have plugged cables in backwards (despite being keyed)
or into the wrong locations. E.g., you have 200 pairs of identical
connectors, who's to say the two desired ends find their way into
the correct connectors? Maybe 1A gets connected to 2B instead of 1B.

If you rely on the cables only being long enough to make the "proper" connections, then you likely have to increase the spacing between
connection points to ensure the "slack" that is necessary in any
connection doesn't allow for the selection of an incorrect destination.

Liz's suggestion of having a default INTERNAL connection has some
merit. That handles the "normal" connections and ANY patch cords
would be exceptions. But, I don't know how that it is practical
given design constraints.

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

Don Y <blockedofcourse@foo.invalid> Wrote in message:r

I have lots of "1-to-1" connections between network appliances.I'm trying to think of the most effective (not efficient!) wayof doing so.E.g., imagine two switch-like appliances (lots of ports arrangedside by side). They have to be connected to each

other.One approach is to locate them physically adjacent and use identicallength cords to connect port 1 to 1, 2 to 2, 3 to 3, etc. This worksas long as 1 ALWAYS connects to 1 and never, e.g., 18!When you have more than two such devices, there are other
options.E.g., imagine a set of 8.You can interleave them: AaBbCcDd and try tot same 1-to-1 connection.With the same caveat.Or, could create "sandwiches" where half of A goes to half of itscounterpart "above" with the other half to the counterpart "
below".Esp for devices with two or more rows of 8P8Cs.You could lump all of the devices of one type together andthose of their counterparts: ABCDabcdOr, not exploit proximity and just put them where they fit: AabcBCDdIts an issue because you have a lot
of potential "tangle" as well asgaining physical access to ports that may end up buried in a ratsnest.The datacenter solution is just to dress the cables off to the sideand keep the connection faces relatively exposed. That comes at theexpense of more
cable and the difficulty of extracting/inserting acable from that neat bundle. (Recall that I have to address userswho may be blind or physically disabled.)

Half of A goes to B, and the other half goes to C.

Looks neater.

Cheers
--


----Android NewsGroup Reader---- https://piaohong.s3-us-west-2.amazonaws.com/usenet/index.html

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

T24gNy8zLzIwMjUgMTA6MTUgQU0sIE1hcnRpbiBSaWQgd3JvdGU6DQo+IERvbiBZIDxibG9j a2Vkb2Zjb3Vyc2VAZm9vLmludmFsaWQ+IFdyb3RlIGluIG1lc3NhZ2U6cg0KPj4gSSBoYXZl IGxvdHMgb2YgIjEtdG8tMSIgY29ubmVjdGlvbnMgYmV0d2VlbiBuZXR3b3JrIGFwcGxpYW5j ZXMuSSdtIHRyeWluZyB0byB0aGluayBvZiB0aGUgbW9zdCBlZmZlY3RpdmUgKG5vdCBlZmZp Y2llbnQhKSB3YXlvZiBkb2luZyBzby5FLmcuLCBpbWFnaW5lIHR3byBzd2l0Y2gtbGlrZSBh cHBsaWFuY2VzIChsb3RzIG9mIHBvcnRzIGFycmFuZ2Vkc2lkZSBieSBzaWRlKS4gIFRoZXkg aGF2ZSB0byBiZSBjb25uZWN0ZWQgdG8gZWFjaCBvdGhlci5PbmUgYXBwcm9hY2ggaXMgdG8g bG9jYXRlIHRoZW0gcGh5c2ljYWxseSBhZGphY2VudCBhbmQgdXNlIGlkZW50aWNhbGxlbmd0 aCBjb3JkcyB0byBjb25uZWN0IHBvcnQgMSB0byAxLCAyIHRvIDIsIDMgdG8gMywgZXRjLiAg VGhpcyB3b3Jrc2FzIGxvbmcgYXMgMSBBTFdBWVMgY29ubmVjdHMgdG8gMSBhbmQgbmV2ZXIs IGUuZy4sIDE4IVdoZW4geW91IGhhdmUgbW9yZSB0aGFuIHR3byBzdWNoIGRldmljZXMsIHRo ZXJlIGFyZSBvdGhlciBvcHRpb25zLkUuZy4sIGltYWdpbmUgYSBzZXQgb2YgOC5Zb3UgY2Fu IGludGVybGVhdmUgdGhlbTogIEFhQmJDY0RkICBhbmQgdHJ5IHRvdCBzYW1lIDEtdG8tMSBj b25uZWN0aW9uLldpdGggdGhlIHNhbWUgY2F2ZWF0Lk9yLCBjb3VsZCBjcmVhdGUgInNhbmR3 aWNoZXMiIHdoZXJlIGhhbGYgb2YgQSBnb2VzIHRvIGhhbGYgb2YgaXRzY291bnRlcnBhcnQg ImFib3ZlIiB3aXRoIHRoZSBvdGhlciBoYWxmIHRvIHRoZSBjb3VudGVycGFydCAiYmVsb3ci LkVzcCBmb3IgZGV2aWNlcyB3aXRoIHR3byBvciBtb3JlIHJvd3Mgb2YgOFA4Q3MuWW91IGNv dWxkIGx1bXAgYWxsIG9mIHRoZSBkZXZpY2VzIG9mIG9uZSB0eXBlIHRvZ2V0aGVyIGFuZHRo b3NlIG9mIHRoZWlyIGNvdW50ZXJwYXJ0czogIEFCQ0RhYmNkT3IsIG5vdCBleHBsb2l0IHBy b3hpbWl0eSBhbmQganVzdCBwdXQgdGhlbSB3aGVyZSB0aGV5IGZpdDogIEFhYmNCQ0RkSXRz IGFuIGlzc3VlIGJlY2F1c2UgeW91IGhhdmUgYSBsb3Qgb2YgcG90ZW50aWFsICJ0YW5nbGUi IGFzIHdlbGwgYXNnYWluaW5nIHBoeXNpY2FsIGFjY2VzcyB0byBwb3J0cyB0aGF0IG1heSBl bmQgdXAgYnVyaWVkIGluIGEgcmF0c25lc3QuVGhlIGRhdGFjZW50ZXIgc29sdXRpb24gaXMg anVzdCB0byBkcmVzcyB0aGUgY2FibGVzIG9mZiB0byB0aGUgc2lkZWFuZCBrZWVwIHRoZSBj b25uZWN0aW9uIGZhY2VzIHJlbGF0aXZlbHkgZXhwb3NlZC4gIFRoYXQgY29tZXMgYXQgdGhl ZXhwZW5zZSBvZiBtb3JlIGNhYmxlIGFuZCB0aGUgZGlmZmljdWx0eSBvZiBleHRyYWN0aW5n L2luc2VydGluZyBhY2FibGUgZnJvbSB0aGF0IG5lYXQgYnVuZGxlLiAgKFJlY2FsbCB0aGF0 IEkgaGF2ZSB0byBhZGRyZXNzIHVzZXJzd2hvIG1heSBiZSBibGluZCBvciBwaHlzaWNhbGx5 IGRpc2FibGVkLikNCj4gDQo+IEhhbGYgb2YgQSBnb2VzIHRvIEIsIGFuZCB0aGUgb3RoZXIg aGFsZiBnb2VzIHRvIEMuDQoNCkhhISAgV0hJQ0ggQSwgQiBhbmQgQz8NCg0KPiBMb29rcyAg bmVhdGVyLg0KDQpBcHBlYXJhbmNlIGlzbid0IHRoZSBpc3N1ZS4gIEFjY2Vzc2liaWxpdHkg YW5kICJjb3JyZWN0bmVzcyIgLS0gYmVpbmcgYWJsZSB0bw0KZXhhbWluZSB0aGUgd2lyaW5n ICh3aXRoIHlvdXIgZXllcyBvciBvdGhlciBzZW5zb3J5IGFwcGFyYXR1cykgYW5kIHZlcmlm eQ0KdGhhdCBldmVyeXRoaW5nIGlzICJhcyBpdCBzaG91bGQgYmUiIGFzIHdlbGwgYXMgYmVp bmcgYWJsZSB0byBtYWtlIGFsdGVyYXRpb25zDQpjb3JyZWN0bHksIHdpdGhvdXQgaGF2aW5n IHRvIGludmVzdCBhIGxvdCBvZiBlZmZvcnQgaW50byB2ZXJpZnlpbmcgeW91J3ZlIGdvdA0K dGhlIHJpZ2h0IHdpcmUsIHRoZSByaWdodCBkZXN0aW5hdGlvbiwgZXRjLg0K

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On 7/3/25 15:08, Don Y wrote:

Next, a reexamination as to whether 8P8Cs are the "right" connectors
(just because data centers use them doesn't make them suitable for
use in other applications!).   The telephone "plugboard" would be
nice (no orientation problems) but may be an issue electrically.

It looks as though a vertical panel with two rows in the same numerical
order, with the 'outs' above their corresponding 'ins' and joined by
very short interconnecting links in the 'normal' configuration would be
the best arrangement.

If by "rows" you mean "columns", I think that is right.

Arrange the connectors to be adjacent with indicators "outboard".
This would make it easy for sighted/unsighted folks to get a quick
assessment of the state of everything:

O [] [] O O [] [] O O [] [] O
O [] [] O O [] [] O O [] [] O
O [] [] O O [] [] O O [] [] O
O [] [] O O [] [] O O [] [] O
O [] [] O O [] [] O O [] [] O

where:
   [] connection point
   O  indicator

 Below that could be a horizontal projecting shelf
with much longer leads  tensioned with weights so that they could be
pulled up and used to make non-standard interconnections.  When released
they would retract neatly out of the way.

I'm assuming any connection of that sort would be infrequent and
temporary.  So, a single (?) overly long patch cord could do the
trick.

As it (and it alone) would skew across the panel, it should be easy
to identify as an exception.  And, follow to each end.

That was the system used in telephone manual switchboards and it worked
very well, with the interconnections being easy to trace and no tangles
of loose leads.

If you could justify the design and special manufacturing costs, you
could go one stage further and make it so that an ''out' and 'in' pair
of sockets on one 'channel' were automatically connected to each other
in the absence of a plug.  That would do away with the short
interconnectors altogether.

That may be possible.  I had planned on making "singletons" that were effectively edge-stackable as the total number (and configuration)
would vary from system to system.  "Standardizing" on some arbitrary packaging (dimensions) would likely lead to complications at installation.

E.g., here, I would arrange things in columns of 36.  Shorter columns
(e.g., 24) would necessitate more columns (i.e., greater width)
while longer columns would make them span longer reaches so the points
at one end were less accessible than the other.

 This method was used by the BBC for their
apparatus racks, so that only the non-standard interconnections needed
external leads.

If standard Post Office Gauge'B' plugs (316-type) and leads could carry
the signals without degradation, there have been millions of standard
ready-made jack strips with break-jacks already manufactured which will
now be lying around as surplus stock.  (Grab them before the government
tries to buy them all back when they realise manual telephone
switchboards are EMP-proof.)

They're (currently) Gbe drops.  So, there is some care needed in their physical termination.  I suspect they will be replaced with CAT6 in
the not too distant future as Gbe is becoming "old hat".  That may
pose some problems if the connection points are located too closely
together
(minimum bend radius)

Of course, the ultimate expression of that idea is to use a 90 degree
xy matrix, a pin at each join, to route any source to any destination.
Add a Z axis multilayer pin to cover multi path source to destination.

Chris

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On 7/5/2025 6:31 AM, chrisq wrote:

Of course, the ultimate expression of that idea is to use a 90 degree
xy matrix, a pin at each join, to route any source to any destination.
Add a Z axis multilayer pin to cover multi path source to destination.

The normal (and nominal) routing is (or can be) to the adjacent
destination. So, it makes sense to optimize that. The exception
can be... exceptional (and, likely, only needed to provide
temporary relief in the event of an equipment failure).

The user *wants* to be able to sort out where things are routed.

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On 7/5/2025 10:43 AM, chrisq wrote:

On 7/5/25 17:45, Don Y wrote:

On 7/5/2025 6:31 AM, chrisq wrote:

Of course, the ultimate expression of that idea is to use a 90 degree
xy matrix, a pin at each join, to route any source to any destination.
Add a Z axis multilayer pin to cover multi path source to destination.

The normal (and nominal) routing is (or can be) to the adjacent
destination.  So, it makes sense to optimize that.  The exception
can be... exceptional (and, likely, only needed to provide
temporary relief in the event of an equipment failure).

The user *wants* to be able to sort out where things are routed.

Not applicable to all, but have used the small 10x10 x-y matrix
pin boards in the past to patch comms lines, eg: null modem
or pin to pin, or combination of such. Visually obvious which pin
goes to what as well. One of those glued between a pair of 25
pin D, in one example.

[My solution for EIA232-ish "conversions" is to build small
(2"x2") adapters that permanently hardwire a connector (of
a specific gender) on one end to another connector (of a
possibly different gender) on the other end. Then, slap a
label on the assembly: "Null Modem", "Null Terminal", "Male
Gender Swap", "Female Gender Swap", "APC UPS", "PLIP", etc.
The thinking being that if you have to sort out how to connect
X and Y *today*, you will likely want to connect them, again,
tomorrow. So, make a specific adapter, document it, and be
done with it. I only use my "patch board" to sort out how
to make that first adapter.]

Here, "visually obvious" is the problem. I need a solution of
which someone who is blind, physically disabled, etc. can make
effective use.

You (likely) can GLANCE at a network switch and see if a
node is inactive (if you assume all would be active).
And, can see if there is a network cable plugged into
that port.

Close your eyes and do it. Or, confine yourself to a wheelchair
that limits your reach or the ease with which you can "navigate"
a panel with hundreds of connections.

If you were told to swap the cables at connections X and Y,
would you be able to do it -- correctly? (no peeking!)
How do you locate X and Y? Hold onto the X cable while
fiddling for the Y?

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On 7/5/25 17:45, Don Y wrote:

On 7/5/2025 6:31 AM, chrisq wrote:

Of course, the ultimate expression of that idea is to use a 90 degree
xy matrix, a pin at each join, to route any source to any destination.
Add a Z axis multilayer pin to cover multi path source to destination.

The normal (and nominal) routing is (or can be) to the adjacent destination.  So, it makes sense to optimize that.  The exception
can be... exceptional (and, likely, only needed to provide
temporary relief in the event of an equipment failure).

The user *wants* to be able to sort out where things are routed.

Not applicable to all, but have used the small 10x10 x-y matrix
pin boards in the past to patch comms lines, eg: null modem
or pin to pin, or combination of such. Visually obvious which pin
goes to what as well. One of those glued between a pair of 25
pin D, in one example.

Chris

.

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)