Rogue network cable... or not?

Daid

[url=http://img808.imageshack.us/i/imag0014l.jpg/][img]http://img808.imageshack.us/img808/7881/imag0014l.jpg[/img][/url]
[url=http://img819.imageshack.us/i/imag0013fe.jpg/][img]http://img819.imageshack.us/img819/8056/imag0013fe.jpg[/img][/url]

This cable was in place before I got there, I asked our ICT department for a more permanent solution. They suggested i would use a longer cable...
WTF: The cable is apparently a feature, not a rogue cable.
WTF: I cannot replace the cable, those cabinets are locked. And I don't want to.

I expect to see something like this: http://thedailywtf.com/Articles/Sponsor-Appreciation,-The-Server-Room-Switch,-Woody,-and-More.aspx#Pic3 soon.

A 0.5Mbps feature. Rocking.

blakeyrat

Holy crap, does it even work? It's kissing those fluorescents.

OzPeter

@blakeyrat said:

Holy crap, does it even work? It's kissing those fluorescents.

Why shouldn't it work? Cat5 (or whatever) Ethernet cable is twisted pair, so induced currents/voltages from nearby sources hitting a specific pair tend to cancel out as they act equally on both wires in the pair. So with differential (between wires in a a pair) rather than absolute (each wire from a fixed reference point) sampling in the electronics and you have pretty good noise immunity. Plus 50 & 100Hz (assuming Europe) is well outside of the frequency band of Ethernet transmissions.

However with that said this particular cabling run constitutes a hazard and if OSHA (or whatever the European equivalent is) saw it they would come down on the company for an unsafe work environment. And somebody in the company should be aware that hazard=$$$liability if knowingly left that way. And that is without even getting into the potential for equipment damage that this run exposes!

blakeyrat

@OzPeter said:

@blakeyrat said:
Holy crap, does it even work? It's kissing those fluorescents.
Why shouldn't it work? Cat5 (or whatever) Ethernet cable is twisted pair, so induced currents/voltages from nearby sources hitting a specific pair tend to cancel out as they act equally on both wires in the pair. So with differential (between wires in a a pair) rather than absolute (each wire from a fixed reference point) sampling in the electronics and you have pretty good noise immunity. Plus 50 & 100Hz (assuming Europe) is well outside of the frequency band of Ethernet transmissions.

You might have "education" and "science" on your side, but my practical experience tells me Cat5 doesn't work worth shit when run next to fluorescents. I spent days re-doing runs from some idiot contractor, doing nothing but re-tying them 3' away from the lights... lo and behold, suddenly the bad runs were good runs.

RichP

Not a WTF. This is simply a rare installation of TCP/IP-over tin cans and a string. The taut cable is necessary for higer data rates.