I'll respond in line to your questions:



when you're talking about signalling, you've got a base voltage, let's say it's 5 volts. (I forget offhand what ethernet uses). A 0 would equal 5 volts on the line, and a 1 would equal +5v, so it would equal 10volts. That way, if it pulses 5 volts, the NIC interprets that as 0, if it's 10 volts, it interprets that as a 1. The voltages can be arbitrary, but I think they were chosen for a specific reason. What reason? I'm not sure.


You are correct-- MAC addresses are burnt onto the hardware, but they can also be "spoofed", ie, changed, by software. Normally, though, it is burnt in. "Mapping" Just means "translating", or "look up"-- each host carries its own ARP table, and it either performs the lookup for host-> IP address there. If it's not located there, it broadcasts out, "WHO HAS THIS IP ADDRESS?" to the network, and the host with that IP responds with its MAC.

When you say it's a "broadcast" network, that's all you can say-- one host contacts the other essentially by saying, "I WANT TO TALK TO MAC ADDRESS XXXXXXX", and that host responds to the system.



Sort of. ARP kind of "spans" the layers-- it's a table that keeps up with the level 2 MAC address, and correlates that to the Layer 3 IP address. Routing is all done at Layer 3. Routing tables accomplish this, by saying, "I know this IP address is accessible through this other router 1.1.1.1" Usually, there's a "default gateway", that unknown IP addresses are routed to.


Thanks. The Cisco curriculum didn't make any sense to me. As a matter of fact, the concept of ports didn't make sense to me until I started working with multiple services on a system, and I started messing around with Port forwarding and IP Masquerading on routers. You'll run into that fun later, probably



Handshaking isn't really like a ping. A ping would look like this:

System 1: I'm up, system 2, are you there?
System 2: Yeah, I'm up.

Where a handshake basically looks like this:

System 1: I'm ready to send data to you... you ready?
System 2: Yeah, I'm ready-- you ready?
System 1: Oh, I'm SO ready. Here it comes!
kind of silly, but it gets the job done


Really, it could be software, or hardware-- it depends on where it's implemented in the system. (Kind of like how you could have an MPEG decoder card, or you could just have a software codec to do it). Usually, it's done in software.

Typically, this doesn't get mentioned much, because we usually work in ASCII. Look up Character sets to get more information on it-- ASCII uses a certain number of bits to represent a character, EBCIDIC uses another amount, and Unicode uses an entirely different amount of characters to represent that same character. This is where, as I understand it, that translation takes place.



There's always the chance that I'm slightly off. I think that a more precise positioning of the Application layer is at the very end of the process, when the TCP/IP stack hands the data to the actual requesting application. My instructor (keep in mind, this was like, 2-3 years ago) didn't really make too big of a distinction. Basically, I look at the Application layer as the "glue" or "pipe" that actually connects the application to the network.


Trust me, it makes a lot more sense once you start plugging in wires. Also, keep in mind that the OSI model is not the actual model used these days-- the TCP/IP stack is what's used for this. The TCP/IP stack is much more simple:

http://www.pku.edu.cn/academic/resea...ml/TC0102.html
There's only 4 layers:

• Application
• Transport
• Internetwork
• Network access layer

In that one, network access layer is the wires and MAC addressing type of stuff. Internetwork is IP addressing and routing. Transport is the TCP/UDP ports. The Application layer is essentially "everything else"-- where you actually "do stuff with the data". Much more simple, because you don't have to deal with the presentation layer, which is often done implicitly.