I've been using this resource for a while, pity it hasn't been updated in over a year.

Make sure you leave time to write the report before taking the exam.

For future reference:
http://www.no-ip.com/getting_started.php
http://help.logmein.com/selfserviceknowledgerenderer?type=FAQ&id=kA030000000DGCkCAO

Since it's difficult for you to monitor the remote machine, I strongly suggest you play with msfpayload on a machine that you can easily monitor and control. If that means setting up a virtual machine running Windows, do it.

If your machine is already exposed to the internet such that you can attack it, then you can easily monitor its state remotely by setting up SSH, or even some remote desktop over VPN. Log in remotely, run your exploit, check to see if it worked.

No. It's not the most important or required, but you should know at least one programming language, be it Perl, C, Python, or whatever else. The point is, you should be able to quickly write and read code so you can modify exploits, write your own tools, etc. Python is just recommended because it's easy to learn, and you can write some powerful scripts relatively quickly.

If you own both computers, then just try it and find out what happens. Part of learning to hack is experimenting and seeing what the results are and interpreting them.

If you're familiar with other programming languages it's easy enough to learn. I think I got a copy of O'Reilly's Learning Python when I first started out, read the first few chapters, and just used the rest for reference. If I remember correctly it was a pretty decent book.

If you're looking for videos, Google's Code University has some lectures: http://code.google.com/edu/languages/google-python-class/index.html

I prefer a clean install. I use it on a VM, and all my customizations are stored in a repository in my local network, and I can quickly git clone them over to the new install. I have scripts that copy old configurations over and setup everything the way I like it. This allows me to quickly get it up to speed with minimal interaction in a matter of minutes.

Certain sections in the course can be done offline. You have three months to schedule your exam after your lab time ends, so you could easily spend an extra three months working on those sections you feel weak in before you take the exam. Things like buffer overflows are good a good example of things you can practice offline. You can download vulnerable virtual machine images made to be broken into for practice during that time as well, there's plenty of material out there to practice on once your lab time ends.

Been seeding since they announced it on #backtrack-linux

A bit of googling would have given you the answer to fix this: http://www.backtrack-linux.org/forums/showthread.php?t=50394

It's showing in public now?

In fact, there are already hundreds of videos on YouTube dedicated to wireless hacking, making watching this one more trouble than it's worth.

He said look around and you'll find different decoders. There's no guarantee that anyone will do it correctly unless you've found the one that actually encoded it.

I've flipped through Advanced Penetration Testing for Highly-Secured Environments. I have some mixed feelings about it. Some things in the book aren't what I'd consider "advanced" (starting an ftp server, basic nmap scans, snmp scans, selecting a text editor ()). Yet there are some gems in there, like setting up your own virtual lab, and bypassing IDSs and firewalls. Other things it barely touches upon (buffer overflow refresher - but never really goes into detail). There's a section on fuzzing, but only covers basic fuzzing.

Maybe it's just me, but when I see "Advanced" I expect something like this: http://www.inguardians.com/research/docs/Skoudis_pentestsecrets.pdf