We'll start our cabling adventure with the oldest and most unwieldy type of cabling out there: coaxial. If this name is ringing some bells, it is probably because it’s the same type of stuff you use to hook up your cable television. You may also have a little experience with this type of cabling if you currently use cable for your Internet service as well. Coaxial cabling is typically thick and heavy, most often wrapped in black or white insulation. It consists of a heavy copper wire which is shielded by an external copper braid, then wrapped in plastic insulation. "Flexible" coaxial cable can also consist of braided copper wire at its core, instead of a single heavy strand.
Coaxial Cable is extremely durable and essentially immune to electromagnetic interference (EMI) and radio-frequency interference (RMI). It is currently the cable of choice for television transmission, and frequently, Internet service. Coaxial cable offers a high bandwidth, making it a great choice for these types of implementations.
In networking, coaxial cable has largely been replaced by twisted pair and fiberoptic. However, it is still seen from time to time, and we all benefit from a good history lesson from time to time
IEEE (Institute of Electrical and Electronic Engineers) is the group typically responsible for setting standards when it comes to electrical communications. They devised a naming convention for networking cable that follows a specific structure. 10base2 is an example of the name of a certain type of coaxial cable. The 10 refers to the cables rated transmission speed (given in megabits). "Base" refers to baseband, and the 2 defines the maximum segment length in hundreds of meters. Now that we have that out of the way, let's look at the main two types that are in the standard.
This is the IEEE standard for Ethernet cabling with a data rate of 10 megabits per second (Mbps) over RG-58 coaxial cable. While the 2 initially denoted a maximum segment (segment being the length of cable between signal repeaters) length of 200 meters, they later discovered that signal degradation was too severe past 185 meters and changed the "rules". (They clearly didn't change the name, or we'd be reading 10Base1.8). This cable type is commonly referred to by it's "slang" term: Thinnet. Being an Ethernet standard, Thinnet was used in networks in typical Ethernet star and bus configurations, and its physical features don't really effect its technological implementation. It supported longer segments and high bandwidth, making it a good choice for, say, connecting floors in a building. It is a bit unwieldy, and these days its incredibly slow speeds have led to it being all but completely replaced.
When we discussed garden hose cabling in the introduction of this article, we were talking about 10Base5, also appropriately known as Thicknet. Probably the only upside this old cable had was its ability to transmit data 500 meters without needing a signal repeater. Other than that, its large size, weight, and hideous visage (as well as the difficulty of "splitting" it) made it an unpopular choice that administrators were very pleased to replace as soon as possible. Network topologies with this cabling were interesting. Since breaking up its segment lengths, in essence, ruined its only good quality, what we call "hubs" weren't really used. Instead, to add a segment to an existing line anywhere along the cable (ie: if a workstation needed to be added near an existing Thicknet line), administrators would instead use "vampire" connectors, which had barbs to puncture the Thicknet in order to "tap in" to it.